Add files using upload-large-folder tool

examples/magicdata-read/cosyvoice/conf/cosyvoice.fromscratch.yaml

@@ -0,0 +1,203 @@
+# set random seed, so that you may reproduce your result.
+__set_seed1: !apply:random.seed [1986]
+__set_seed2: !apply:numpy.random.seed [1986]
+__set_seed3: !apply:torch.manual_seed [1986]
+__set_seed4: !apply:torch.cuda.manual_seed_all [1986]
+
+# fixed params
+sample_rate: 22050
+text_encoder_input_size: 512
+llm_input_size: 1024
+llm_output_size: 1024
+spk_embed_dim: 192
+
+# model params
+# for all class/function included in this repo, we use !<name> or !<new> for intialization, so that user may find all corresponding class/function according to one single yaml.
+# for system/third_party class/function, we do not require this.
+llm: !new:cosyvoice.llm.llm.TransformerLM
+    text_encoder_input_size: !ref <text_encoder_input_size>
+    llm_input_size: !ref <llm_input_size>
+    llm_output_size: !ref <llm_output_size>
+    text_token_size: 51866 # change to 60515 if you want to train with CosyVoice-300M-25Hz recipe
+    speech_token_size: 4096
+    length_normalized_loss: True
+    lsm_weight: 0
+    spk_embed_dim: !ref <spk_embed_dim>
+    text_encoder: !new:cosyvoice.transformer.encoder.ConformerEncoder
+        input_size: !ref <text_encoder_input_size>
+        output_size: 1024
+        attention_heads: 8
+        linear_units: 2048
+        num_blocks: 3
+        dropout_rate: 0.1
+        positional_dropout_rate: 0.1
+        attention_dropout_rate: 0.0
+        normalize_before: True
+        input_layer: 'linear'
+        pos_enc_layer_type: 'rel_pos_espnet'
+        selfattention_layer_type: 'rel_selfattn'
+        use_cnn_module: False
+        macaron_style: False
+        use_dynamic_chunk: False
+        use_dynamic_left_chunk: False
+        static_chunk_size: 1
+    llm: !new:cosyvoice.transformer.encoder.TransformerEncoder
+        input_size: !ref <llm_input_size>
+        output_size: !ref <llm_output_size>
+        attention_heads: 8
+        linear_units: 2048
+        num_blocks: 7
+        dropout_rate: 0.1
+        positional_dropout_rate: 0.1
+        attention_dropout_rate: 0.0
+        input_layer: 'linear_legacy'
+        pos_enc_layer_type: 'rel_pos_espnet'
+        selfattention_layer_type: 'rel_selfattn'
+        static_chunk_size: 1
+    sampling: !name:cosyvoice.utils.common.ras_sampling
+        top_p: 0.8
+        top_k: 25
+        win_size: 10
+        tau_r: 0.1
+
+flow: !new:cosyvoice.flow.flow.MaskedDiffWithXvec
+    input_size: 512
+    output_size: 80
+    spk_embed_dim: !ref <spk_embed_dim>
+    output_type: 'mel'
+    vocab_size: 4096
+    input_frame_rate: 50 # change to 25 if you want to train with CosyVoice-300M-25Hz recipe
+    only_mask_loss: True
+    encoder: !new:cosyvoice.transformer.encoder.ConformerEncoder
+        output_size: 512
+        attention_heads: 4
+        linear_units: 1024
+        num_blocks: 3
+        dropout_rate: 0.1
+        positional_dropout_rate: 0.1
+        attention_dropout_rate: 0.1
+        normalize_before: True
+        input_layer: 'linear'
+        pos_enc_layer_type: 'rel_pos_espnet'
+        selfattention_layer_type: 'rel_selfattn'
+        input_size: 512
+        use_cnn_module: False
+        macaron_style: False
+    length_regulator: !new:cosyvoice.flow.length_regulator.InterpolateRegulator
+        channels: 80
+        sampling_ratios: [1, 1, 1, 1]
+    decoder: !new:cosyvoice.flow.flow_matching.ConditionalCFM
+        in_channels: 240
+        n_spks: 1
+        spk_emb_dim: 80
+        cfm_params: !new:omegaconf.DictConfig
+            content:
+                sigma_min: 1e-06
+                solver: 'euler'
+                t_scheduler: 'cosine'
+                training_cfg_rate: 0.2
+                inference_cfg_rate: 0.7
+                reg_loss_type: 'l1'
+        estimator: !new:cosyvoice.flow.decoder.ConditionalDecoder
+            in_channels: 320
+            out_channels: 80
+            channels: [256, 256]
+            dropout: 0.0
+            attention_head_dim: 64
+            n_blocks: 4
+            num_mid_blocks: 8
+            num_heads: 8
+            act_fn: 'gelu'
+
+hift: !new:cosyvoice.hifigan.generator.HiFTGenerator
+    in_channels: 80
+    base_channels: 512
+    nb_harmonics: 8
+    sampling_rate: !ref <sample_rate>
+    nsf_alpha: 0.1
+    nsf_sigma: 0.003
+    nsf_voiced_threshold: 10
+    upsample_rates: [8, 8]
+    upsample_kernel_sizes: [16, 16]
+    istft_params:
+        n_fft: 16
+        hop_len: 4
+    resblock_kernel_sizes: [3, 7, 11]
+    resblock_dilation_sizes: [[1, 3, 5], [1, 3, 5], [1, 3, 5]]
+    source_resblock_kernel_sizes: [7, 11]
+    source_resblock_dilation_sizes: [[1, 3, 5], [1, 3, 5]]
+    lrelu_slope: 0.1
+    audio_limit: 0.99
+    f0_predictor: !new:cosyvoice.hifigan.f0_predictor.ConvRNNF0Predictor
+        num_class: 1
+        in_channels: 80
+        cond_channels: 512
+
+# processor functions
+parquet_opener: !name:cosyvoice.dataset.processor.parquet_opener
+get_tokenizer: !name:whisper.tokenizer.get_tokenizer # change to !name:cosyvoice.tokenizer.tokenizer.get_tokenizer if you want to train with CosyVoice-300M-25Hz recipe
+    multilingual: True
+    num_languages: 100
+    language: 'en'
+    task: 'transcribe'
+allowed_special: 'all'
+tokenize: !name:cosyvoice.dataset.processor.tokenize
+    get_tokenizer: !ref <get_tokenizer>
+    allowed_special: !ref <allowed_special>
+filter: !name:cosyvoice.dataset.processor.filter
+    max_length: 40960
+    min_length: 0
+    token_max_length: 200
+    token_min_length: 1
+resample: !name:cosyvoice.dataset.processor.resample
+    resample_rate: !ref <sample_rate>
+feat_extractor: !name:matcha.utils.audio.mel_spectrogram
+    n_fft: 1024
+    num_mels: 80
+    sampling_rate: !ref <sample_rate>
+    hop_size: 256
+    win_size: 1024
+    fmin: 0
+    fmax: 8000
+    center: False
+compute_fbank: !name:cosyvoice.dataset.processor.compute_fbank
+    feat_extractor: !ref <feat_extractor>
+parse_embedding: !name:cosyvoice.dataset.processor.parse_embedding
+    normalize: True
+shuffle: !name:cosyvoice.dataset.processor.shuffle
+    shuffle_size: 1000
+sort: !name:cosyvoice.dataset.processor.sort
+    sort_size: 500  # sort_size should be less than shuffle_size
+batch: !name:cosyvoice.dataset.processor.batch
+    batch_type: 'dynamic'
+    max_frames_in_batch: 12000
+padding: !name:cosyvoice.dataset.processor.padding
+    use_spk_embedding: False # change to True during sft
+
+# dataset processor pipeline
+data_pipeline: [
+    !ref <parquet_opener>,
+    !ref <tokenize>,
+    !ref <filter>,
+    !ref <resample>,
+    !ref <compute_fbank>,
+    !ref <parse_embedding>,
+    !ref <shuffle>,
+    !ref <sort>,
+    !ref <batch>,
+    !ref <padding>,
+]
+
+# train conf
+train_conf:
+    optim: adam
+    optim_conf:
+        lr: 0.002 # change to 0.001 if you want to train flow from scratch
+    scheduler: warmuplr
+    scheduler_conf:
+        warmup_steps: 25000
+    max_epoch: 200
+    grad_clip: 5
+    accum_grad: 2
+    log_interval: 100
+    save_per_step: -1

examples/magicdata-read/cosyvoice/conf/cosyvoice.yaml

@@ -0,0 +1,203 @@
+# set random seed, so that you may reproduce your result.
+__set_seed1: !apply:random.seed [1986]
+__set_seed2: !apply:numpy.random.seed [1986]
+__set_seed3: !apply:torch.manual_seed [1986]
+__set_seed4: !apply:torch.cuda.manual_seed_all [1986]
+
+# fixed params
+sample_rate: 22050
+text_encoder_input_size: 512
+llm_input_size: 1024
+llm_output_size: 1024
+spk_embed_dim: 192
+
+# model params
+# for all class/function included in this repo, we use !<name> or !<new> for intialization, so that user may find all corresponding class/function according to one single yaml.
+# for system/third_party class/function, we do not require this.
+llm: !new:cosyvoice.llm.llm.TransformerLM
+    text_encoder_input_size: !ref <text_encoder_input_size>
+    llm_input_size: !ref <llm_input_size>
+    llm_output_size: !ref <llm_output_size>
+    text_token_size: 51866 # change to 60515 if you want to train with CosyVoice-300M-25Hz recipe
+    speech_token_size: 4096
+    length_normalized_loss: True
+    lsm_weight: 0
+    spk_embed_dim: !ref <spk_embed_dim>
+    text_encoder: !new:cosyvoice.transformer.encoder.ConformerEncoder
+        input_size: !ref <text_encoder_input_size>
+        output_size: 1024
+        attention_heads: 16
+        linear_units: 4096
+        num_blocks: 6
+        dropout_rate: 0.1
+        positional_dropout_rate: 0.1
+        attention_dropout_rate: 0.0
+        normalize_before: True
+        input_layer: 'linear'
+        pos_enc_layer_type: 'rel_pos_espnet'
+        selfattention_layer_type: 'rel_selfattn'
+        use_cnn_module: False
+        macaron_style: False
+        use_dynamic_chunk: False
+        use_dynamic_left_chunk: False
+        static_chunk_size: 1
+    llm: !new:cosyvoice.transformer.encoder.TransformerEncoder
+        input_size: !ref <llm_input_size>
+        output_size: !ref <llm_output_size>
+        attention_heads: 16
+        linear_units: 4096
+        num_blocks: 14
+        dropout_rate: 0.1
+        positional_dropout_rate: 0.1
+        attention_dropout_rate: 0.0
+        input_layer: 'linear_legacy'
+        pos_enc_layer_type: 'rel_pos_espnet'
+        selfattention_layer_type: 'rel_selfattn'
+        static_chunk_size: 1
+    sampling: !name:cosyvoice.utils.common.ras_sampling
+        top_p: 0.8
+        top_k: 25
+        win_size: 10
+        tau_r: 0.1
+
+flow: !new:cosyvoice.flow.flow.MaskedDiffWithXvec
+    input_size: 512
+    output_size: 80
+    spk_embed_dim: !ref <spk_embed_dim>
+    output_type: 'mel'
+    vocab_size: 4096
+    input_frame_rate: 50 # change to 25 if you want to train with CosyVoice-300M-25Hz recipe
+    only_mask_loss: True
+    encoder: !new:cosyvoice.transformer.encoder.ConformerEncoder
+        output_size: 512
+        attention_heads: 8
+        linear_units: 2048
+        num_blocks: 6
+        dropout_rate: 0.1
+        positional_dropout_rate: 0.1
+        attention_dropout_rate: 0.1
+        normalize_before: True
+        input_layer: 'linear'
+        pos_enc_layer_type: 'rel_pos_espnet'
+        selfattention_layer_type: 'rel_selfattn'
+        input_size: 512
+        use_cnn_module: False
+        macaron_style: False
+    length_regulator: !new:cosyvoice.flow.length_regulator.InterpolateRegulator
+        channels: 80
+        sampling_ratios: [1, 1, 1, 1]
+    decoder: !new:cosyvoice.flow.flow_matching.ConditionalCFM
+        in_channels: 240
+        n_spks: 1
+        spk_emb_dim: 80
+        cfm_params: !new:omegaconf.DictConfig
+            content:
+                sigma_min: 1e-06
+                solver: 'euler'
+                t_scheduler: 'cosine'
+                training_cfg_rate: 0.2
+                inference_cfg_rate: 0.7
+                reg_loss_type: 'l1'
+        estimator: !new:cosyvoice.flow.decoder.ConditionalDecoder
+            in_channels: 320
+            out_channels: 80
+            channels: [256, 256]
+            dropout: 0.0
+            attention_head_dim: 64
+            n_blocks: 4
+            num_mid_blocks: 12
+            num_heads: 8
+            act_fn: 'gelu'
+
+hift: !new:cosyvoice.hifigan.generator.HiFTGenerator
+    in_channels: 80
+    base_channels: 512
+    nb_harmonics: 8
+    sampling_rate: !ref <sample_rate>
+    nsf_alpha: 0.1
+    nsf_sigma: 0.003
+    nsf_voiced_threshold: 10
+    upsample_rates: [8, 8]
+    upsample_kernel_sizes: [16, 16]
+    istft_params:
+        n_fft: 16
+        hop_len: 4
+    resblock_kernel_sizes: [3, 7, 11]
+    resblock_dilation_sizes: [[1, 3, 5], [1, 3, 5], [1, 3, 5]]
+    source_resblock_kernel_sizes: [7, 11]
+    source_resblock_dilation_sizes: [[1, 3, 5], [1, 3, 5]]
+    lrelu_slope: 0.1
+    audio_limit: 0.99
+    f0_predictor: !new:cosyvoice.hifigan.f0_predictor.ConvRNNF0Predictor
+        num_class: 1
+        in_channels: 80
+        cond_channels: 512
+
+# processor functions
+parquet_opener: !name:cosyvoice.dataset.processor.parquet_opener
+get_tokenizer: !name:whisper.tokenizer.get_tokenizer # change to !name:cosyvoice.tokenizer.tokenizer.get_tokenizer if you want to train with CosyVoice-300M-25Hz recipe
+    multilingual: True
+    num_languages: 100
+    language: 'en'
+    task: 'transcribe'
+allowed_special: 'all'
+tokenize: !name:cosyvoice.dataset.processor.tokenize
+    get_tokenizer: !ref <get_tokenizer>
+    allowed_special: !ref <allowed_special>
+filter: !name:cosyvoice.dataset.processor.filter
+    max_length: 40960
+    min_length: 0
+    token_max_length: 200
+    token_min_length: 1
+resample: !name:cosyvoice.dataset.processor.resample
+    resample_rate: !ref <sample_rate>
+feat_extractor: !name:matcha.utils.audio.mel_spectrogram
+    n_fft: 1024
+    num_mels: 80
+    sampling_rate: !ref <sample_rate>
+    hop_size: 256
+    win_size: 1024
+    fmin: 0
+    fmax: 8000
+    center: False
+compute_fbank: !name:cosyvoice.dataset.processor.compute_fbank
+    feat_extractor: !ref <feat_extractor>
+parse_embedding: !name:cosyvoice.dataset.processor.parse_embedding
+    normalize: True
+shuffle: !name:cosyvoice.dataset.processor.shuffle
+    shuffle_size: 1000
+sort: !name:cosyvoice.dataset.processor.sort
+    sort_size: 500  # sort_size should be less than shuffle_size
+batch: !name:cosyvoice.dataset.processor.batch
+    batch_type: 'dynamic'
+    max_frames_in_batch: 2000
+padding: !name:cosyvoice.dataset.processor.padding
+    use_spk_embedding: False # change to True during sft
+
+# dataset processor pipeline
+data_pipeline: [
+    !ref <parquet_opener>,
+    !ref <tokenize>,
+    !ref <filter>,
+    !ref <resample>,
+    !ref <compute_fbank>,
+    !ref <parse_embedding>,
+    !ref <shuffle>,
+    !ref <sort>,
+    !ref <batch>,
+    !ref <padding>,
+]
+
+# train conf
+train_conf:
+    optim: adam
+    optim_conf:
+        lr: 0.001 # change to 1e-5 during sft
+    scheduler: warmuplr # change to constantlr during sft
+    scheduler_conf:
+        warmup_steps: 2500
+    max_epoch: 200
+    grad_clip: 5
+    accum_grad: 2
+    log_interval: 100
+    save_per_step: -1

examples/magicdata-read/cosyvoice/conf/ds_stage2.json

@@ -0,0 +1,42 @@
+{
+  "train_micro_batch_size_per_gpu": 1,
+  "gradient_accumulation_steps": 1,
+  "steps_per_print": 100,
+  "gradient_clipping": 5,
+  "fp16": {
+    "enabled": false,
+    "auto_cast": false,
+    "loss_scale": 0,
+    "initial_scale_power": 16,
+    "loss_scale_window": 256,
+    "hysteresis": 2,
+    "consecutive_hysteresis": false,
+    "min_loss_scale": 1
+  },
+  "bf16": {
+    "enabled": false
+  },
+  "zero_force_ds_cpu_optimizer": false,
+  "zero_optimization": {
+    "stage": 2,
+    "offload_optimizer": {
+      "device": "none",
+      "pin_memory": true
+    },
+    "allgather_partitions": true,
+    "allgather_bucket_size": 5e8,
+    "overlap_comm": false,
+    "reduce_scatter": true,
+    "reduce_bucket_size": 5e8,
+    "contiguous_gradients" : true
+  },
+  "optimizer": {
+    "type": "AdamW",
+    "params": {
+        "lr": 0.001,
+        "weight_decay": 0.0001,
+        "torch_adam": true,
+        "adam_w_mode": true
+    }
+  }
+}

examples/magicdata-read/cosyvoice/local/download_and_untar.sh

@@ -0,0 +1,97 @@
+#!/bin/bash
+
+# Copyright   2014  Johns Hopkins University (author: Daniel Povey)
+# Apache 2.0
+
+remove_archive=false
+
+if [ "$1" == --remove-archive ]; then
+  remove_archive=true
+  shift
+fi
+
+if [ $# -ne 3 ]; then
+  echo "Usage: $0 [--remove-archive] <data-base> <url-base> <corpus-part>"
+  echo "e.g.: $0 /export/a15/vpanayotov/data www.openslr.org/resources/11 dev-clean"
+  echo "With --remove-archive it will remove the archive after successfully un-tarring it."
+  echo "<corpus-part> can be one of: dev-clean, test-clean, dev-other, test-other,"
+  echo "          train-clean-100, train-clean-360, train-other-500."
+  exit 1
+fi
+
+data=$1
+url=$2
+part=$3
+
+if [ ! -d "$data" ]; then
+  echo "$0: no such directory $data"
+  exit 1
+fi
+
+part_ok=false
+list="dev_set test_set train_set"
+for x in $list; do
+  if [ "$part" == $x ]; then part_ok=true; fi
+done
+if ! $part_ok; then
+  echo "$0: expected <corpus-part> to be one of $list, but got '$part'"
+  exit 1
+fi
+
+if [ -z "$url" ]; then
+  echo "$0: empty URL base."
+  exit 1
+fi
+
+if [ -f $data/.$part.complete ]; then
+  echo "$0: data part $part was already successfully extracted, nothing to do."
+  exit 0
+fi
+
+
+# sizes of the archive files in bytes.  This is some older versions.
+sizes_old="1035537823 2201936013 52627842921"
+# sizes_new is the archive file sizes of the final release.  Some of these sizes are of
+# things we probably won't download.
+sizes_new="3886385"
+
+if [ -f $data/$part.tar.gz ]; then
+  size=$(/bin/ls -l $data/$part.tar.gz | awk '{print $5}')
+  size_ok=false
+  for s in $sizes_old $sizes_new; do if [ $s == $size ]; then size_ok=true; fi; done
+  if ! $size_ok; then
+    echo "$0: removing existing file $data/$part.tar.gz because its size in bytes $size"
+    echo "does not equal the size of one of the archives."
+    rm $data/$part.tar.gz
+  else
+    echo "$data/$part.tar.gz exists and appears to be complete."
+  fi
+fi
+
+if [ ! -f $data/$part.tar.gz ]; then
+  if ! which wget >/dev/null; then
+    echo "$0: wget is not installed."
+    exit 1
+  fi
+  full_url=$url/$part.tar.gz
+  echo "$0: downloading data from $full_url.  This may take some time, please be patient."
+
+  if ! wget -P $data --no-check-certificate $full_url; then
+    echo "$0: error executing wget $full_url"
+    exit 1
+  fi
+fi
+
+if ! tar -C $data -xvzf $data/$part.tar.gz; then
+  echo "$0: error un-tarring archive $data/$part.tar.gz"
+  exit 1
+fi
+
+touch $data/.$part.complete
+
+echo "$0: Successfully downloaded and un-tarred $data/$part.tar.gz"
+
+if $remove_archive; then
+  echo "$0: removing $data/$part.tar.gz file since --remove-archive option was supplied."
+  rm $data/$part.tar.gz
+fi

examples/magicdata-read/cosyvoice/local/prepare_data.py

@@ -0,0 +1,52 @@
+import argparse
+import logging
+import os
+from tqdm import tqdm
+
+
+logger = logging.getLogger()
+
+
+def main():
+    utt2wav, utt2text, utt2spk, spk2utt = {}, {}, {}, {}
+    with open(os.path.join(args.src_dir, "TRANS.txt"), "r") as f:
+        lines = f.readlines()[1:]
+        lines = [l.split('\t') for l in lines]
+    for wav, spk, content in tqdm(lines):
+        wav, spk, content = wav.strip(), spk.strip(), content.strip()
+        content = content.replace('[FIL]', '')
+        content = content.replace('[SPK]', '')
+        wav = os.path.join(args.src_dir, spk, wav)
+        if not os.path.exists(wav):
+            continue
+        utt = os.path.basename(wav).replace('.wav', '')
+        utt2wav[utt] = wav
+        utt2text[utt] = content
+        utt2spk[utt] = spk
+        if spk not in spk2utt:
+            spk2utt[spk] = []
+        spk2utt[spk].append(utt)
+
+    with open('{}/wav.scp'.format(args.des_dir), 'w') as f:
+        for k, v in utt2wav.items():
+            f.write('{} {}\n'.format(k, v))
+    with open('{}/text'.format(args.des_dir), 'w') as f:
+        for k, v in utt2text.items():
+            f.write('{} {}\n'.format(k, v))
+    with open('{}/utt2spk'.format(args.des_dir), 'w') as f:
+        for k, v in utt2spk.items():
+            f.write('{} {}\n'.format(k, v))
+    with open('{}/spk2utt'.format(args.des_dir), 'w') as f:
+        for k, v in spk2utt.items():
+            f.write('{} {}\n'.format(k, ' '.join(v)))
+    return
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument('--src_dir',
+                        type=str)
+    parser.add_argument('--des_dir',
+                        type=str)
+    args = parser.parse_args()
+    main()

examples/magicdata-read/cosyvoice/tts_text.json

@@ -0,0 +1,18 @@
+{
+  "38_5718_20170915093303": [
+    "我想这出最好歌曲把歌词发到网上请别人帮我作曲急急",
+    "叫他明天早上差五分儿九点去机场"
+  ],
+  "38_5721_20170915091235": [
+    "变温室调到零下两度档",
+    "交谈中请勿轻信汇款信息陌生电话请勿使用外挂软件"
+  ],
+  "38_5733_20170915130323": [
+    "这是老鹰乐队的一首经典歌曲",
+    "我急用这段音乐我自己找到一段但是有现场杂音"
+  ],
+  "38_5836_20170916221414": [
+    "给我播一个陶喆的专辑",
+    "这套餐好贵呀我发这么多短信贵死了"
+  ]
+}

runtime/python/Dockerfile

@@ -0,0 +1,13 @@
+FROM pytorch/pytorch:2.0.1-cuda11.7-cudnn8-runtime
+ENV DEBIAN_FRONTEND=noninteractive
+
+WORKDIR /opt/CosyVoice
+
+RUN sed -i s@/archive.ubuntu.com/@/mirrors.aliyun.com/@g /etc/apt/sources.list
+RUN apt-get update -y
+RUN apt-get -y install git unzip git-lfs
+RUN git lfs install
+RUN git clone --recursive https://github.com/FunAudioLLM/CosyVoice.git
+# here we use python==3.10 because we cannot find an image which have both python3.8 and torch2.0.1-cu118 installed
+RUN cd CosyVoice && pip3 install -r requirements.txt -i https://mirrors.aliyun.com/pypi/simple/ --trusted-host=mirrors.aliyun.com
+RUN cd CosyVoice/runtime/python/grpc && python3 -m grpc_tools.protoc -I. --python_out=. --grpc_python_out=. cosyvoice.proto

runtime/python/fastapi/client.py

@@ -0,0 +1,92 @@
+# Copyright (c) 2024 Alibaba Inc (authors: Xiang Lyu)
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#   http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+import argparse
+import logging
+import requests
+import torch
+import torchaudio
+import numpy as np
+
+
+def main():
+    url = "http://{}:{}/inference_{}".format(args.host, args.port, args.mode)
+    if args.mode == 'sft':
+        payload = {
+            'tts_text': args.tts_text,
+            'spk_id': args.spk_id
+        }
+        response = requests.request("GET", url, data=payload, stream=True)
+    elif args.mode == 'zero_shot':
+        payload = {
+            'tts_text': args.tts_text,
+            'prompt_text': args.prompt_text
+        }
+        files = [('prompt_wav', ('prompt_wav', open(args.prompt_wav, 'rb'), 'application/octet-stream'))]
+        response = requests.request("GET", url, data=payload, files=files, stream=True)
+    elif args.mode == 'cross_lingual':
+        payload = {
+            'tts_text': args.tts_text,
+        }
+        files = [('prompt_wav', ('prompt_wav', open(args.prompt_wav, 'rb'), 'application/octet-stream'))]
+        response = requests.request("GET", url, data=payload, files=files, stream=True)
+    else:
+        payload = {
+            'tts_text': args.tts_text,
+            'spk_id': args.spk_id,
+            'instruct_text': args.instruct_text
+        }
+        response = requests.request("GET", url, data=payload, stream=True)
+    tts_audio = b''
+    for r in response.iter_content(chunk_size=16000):
+        tts_audio += r
+    tts_speech = torch.from_numpy(np.array(np.frombuffer(tts_audio, dtype=np.int16))).unsqueeze(dim=0)
+    logging.info('save response to {}'.format(args.tts_wav))
+    torchaudio.save(args.tts_wav, tts_speech, target_sr)
+    logging.info('get response')
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument('--host',
+                        type=str,
+                        default='0.0.0.0')
+    parser.add_argument('--port',
+                        type=int,
+                        default='50000')
+    parser.add_argument('--mode',
+                        default='sft',
+                        choices=['sft', 'zero_shot', 'cross_lingual', 'instruct'],
+                        help='request mode')
+    parser.add_argument('--tts_text',
+                        type=str,
+                        default='你好，我是通义千问语音合成大模型，请问有什么可以帮您的吗？')
+    parser.add_argument('--spk_id',
+                        type=str,
+                        default='中文女')
+    parser.add_argument('--prompt_text',
+                        type=str,
+                        default='希望你以后能够做的比我还好呦。')
+    parser.add_argument('--prompt_wav',
+                        type=str,
+                        default='../../../asset/zero_shot_prompt.wav')
+    parser.add_argument('--instruct_text',
+                        type=str,
+                        default='Theo \'Crimson\', is a fiery, passionate rebel leader. \
+                                 Fights with fervor for justice, but struggles with impulsiveness.')
+    parser.add_argument('--tts_wav',
+                        type=str,
+                        default='demo.wav')
+    args = parser.parse_args()
+    prompt_sr, target_sr = 16000, 22050
+    main()

runtime/python/fastapi/server.py

@@ -0,0 +1,171 @@
+# Copyright (c) 2024 Alibaba Inc (authors: Xiang Lyu)
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#   http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+import os
+import sys
+import argparse
+import logging
+logging.getLogger('matplotlib').setLevel(logging.WARNING)
+from fastapi import FastAPI, UploadFile, Form, File
+from fastapi.responses import StreamingResponse
+from fastapi.responses import Response
+from fastapi.middleware.cors import CORSMiddleware
+import uvicorn
+import numpy as np
+ROOT_DIR = os.path.dirname(os.path.abspath(__file__))
+sys.path.append('{}/../../..'.format(ROOT_DIR))
+sys.path.append('{}/../../../third_party/Matcha-TTS'.format(ROOT_DIR))
+from cosyvoice.cli.cosyvoice import CosyVoice, CosyVoice2
+#from cosyvoice.utils.file_utils import load_wav
+
+from fastapi import HTTPException
+import requests
+import tempfile
+import torchaudio
+
+# Keep the original load_wav function unchanged
+def load_wav(wav, target_sr):
+    speech, sample_rate = torchaudio.load(wav, backend='soundfile')
+    speech = speech.mean(dim=0, keepdim=True)
+    if sample_rate != target_sr:
+        assert sample_rate > target_sr, 'wav sample rate {} must be greater than {}'.format(sample_rate, target_sr)
+        speech = torchaudio.transforms.Resample(orig_freq=sample_rate, new_freq=target_sr)(speech)
+    return speech
+
+# Add a new function to handle URLs
+def load_wav_from_url(url, target_sr):
+    # Download the file from the URL to a temporary file
+    response = requests.get(url)
+    if response.status_code != 200:
+        raise HTTPException(status_code=400, detail=f"Failed to download audio from URL: {url}")
+    
+    # Create a temporary file
+    with tempfile.NamedTemporaryFile(suffix='.wav', delete=False) as temp_file:
+        temp_file.write(response.content)
+        temp_file.flush()
+        temp_path = temp_file.name
+    
+    try:
+        # Use the existing load_wav function with the file path
+        speech, sample_rate = torchaudio.load(temp_path, backend='soundfile')
+        speech = speech.mean(dim=0, keepdim=True)
+        if sample_rate != target_sr:
+            assert sample_rate > target_sr, 'wav sample rate {} must be greater than {}'.format(sample_rate, target_sr)
+            speech = torchaudio.transforms.Resample(orig_freq=sample_rate, new_freq=target_sr)(speech)
+        return speech
+    finally:
+        # Clean up the temporary file
+        os.unlink(temp_path)
+
+app = FastAPI()
+# set cross region allowance
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"])
+
+
+def generate_data(model_output):
+    for i in model_output:
+        tts_audio = (i['tts_speech'].numpy() * (2 ** 15)).astype(np.int16).tobytes()
+        yield tts_audio
+
+
+@app.get("/inference_sft")
+@app.post("/inference_sft")
+async def inference_sft(tts_text: str = Form(), spk_id: str = Form()):
+    model_output = cosyvoice.inference_sft(tts_text, spk_id)
+    return StreamingResponse(generate_data(model_output))
+
+
+@app.get("/inference_zero_shot")
+@app.post("/inference_zero_shot")
+async def inference_zero_shot(
+    tts_text: str = Form(), 
+    prompt_text: str = Form(), 
+    prompt_wav_url: str = Form(...),  # Using ... makes this parameter required
+    speed: float = Form(...)
+):
+    # Process the URL directly - no need for conditionals
+    prompt_speech_16k = load_wav_from_url(prompt_wav_url, 16000)
+    
+    # Rest of the function remains the same
+    model_output = cosyvoice.inference_zero_shot(tts_text, prompt_text, prompt_speech_16k, stream=False, speed=speed)
+    
+    # Collect all audio data instead of streaming it
+    audio_data = bytearray()
+    for chunk in generate_data(model_output):
+        audio_data.extend(chunk)
+        
+    # Return complete audio file
+    return Response(
+        content=bytes(audio_data),
+        media_type="audio/wav",
+        headers={"Content-Disposition": "attachment; filename=tts_output.wav"}
+    )
+
+@app.get("/inference_cross_lingual")
+@app.post("/inference_cross_lingual")
+async def inference_cross_lingual(tts_text: str = Form(), prompt_wav: UploadFile = File()):
+    prompt_speech_16k = load_wav(prompt_wav.file, 16000)
+    model_output = cosyvoice.inference_cross_lingual(tts_text, prompt_speech_16k)
+    return StreamingResponse(generate_data(model_output))
+
+
+@app.get("/inference_instruct")
+@app.post("/inference_instruct")
+async def inference_instruct(tts_text: str = Form(), spk_id: str = Form(), instruct_text: str = Form()):
+    model_output = cosyvoice.inference_instruct(tts_text, spk_id, instruct_text)
+    return StreamingResponse(generate_data(model_output))
+
+@app.get("/inference_instruct2")
+@app.post("/inference_instruct2")
+async def inference_instruct2(tts_text: str = Form(), instruct_text: str = Form(), prompt_wav: UploadFile = File(), speed: float = Form(...)):
+    prompt_speech_16k = load_wav(prompt_wav.file, 16000)
+    
+    # Disable streaming by setting stream=False (assuming the function accepts this parameter)
+    model_output = cosyvoice.inference_instruct2(tts_text, instruct_text, prompt_speech_16k, stream=False, speed=speed)
+    
+    # Collect all audio data instead of streaming it
+    audio_data = bytearray()
+    for chunk in generate_data(model_output):
+        audio_data.extend(chunk)
+    print("instruct模式生成成功！")
+    # Return complete audio file
+    return Response(
+        content=bytes(audio_data),
+        media_type="audio/wav",
+        headers={"Content-Disposition": "attachment; filename=tts_output.wav"}
+    )
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser()
+    parser.add_argument('--port',
+                        type=int,
+                        default=50000)
+    parser.add_argument('--model_dir',
+                        type=str,
+                        default='pretrained_models/CosyVoice2-0.5B',
+                        help='local path or modelscope repo id')
+    args = parser.parse_args()
+    #try:
+    #    cosyvoice = CosyVoice(args.model_dir)
+    #except Exception:
+    try:
+        #cosyvoice = CosyVoice2(args.model_dir, load_jit=False, load_trt=True, fp16=True)
+        cosyvoice = CosyVoice2(args.model_dir, load_jit=False, load_trt=True, fp16=True)
+    except Exception:
+        raise TypeError('no valid model_type!')
+    uvicorn.run(app, host="0.0.0.0", port=8000)

runtime/python/grpc/client.py

@@ -0,0 +1,106 @@
+# Copyright (c) 2024 Alibaba Inc (authors: Xiang Lyu)
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#   http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+import os
+import sys
+ROOT_DIR = os.path.dirname(os.path.abspath(__file__))
+sys.path.append('{}/../../..'.format(ROOT_DIR))
+sys.path.append('{}/../../../third_party/Matcha-TTS'.format(ROOT_DIR))
+import logging
+import argparse
+import torchaudio
+import cosyvoice_pb2
+import cosyvoice_pb2_grpc
+import grpc
+import torch
+import numpy as np
+from cosyvoice.utils.file_utils import load_wav
+
+
+def main():
+    with grpc.insecure_channel("{}:{}".format(args.host, args.port)) as channel:
+        stub = cosyvoice_pb2_grpc.CosyVoiceStub(channel)
+        request = cosyvoice_pb2.Request()
+        if args.mode == 'sft':
+            logging.info('send sft request')
+            sft_request = cosyvoice_pb2.sftRequest()
+            sft_request.spk_id = args.spk_id
+            sft_request.tts_text = args.tts_text
+            request.sft_request.CopyFrom(sft_request)
+        elif args.mode == 'zero_shot':
+            logging.info('send zero_shot request')
+            zero_shot_request = cosyvoice_pb2.zeroshotRequest()
+            zero_shot_request.tts_text = args.tts_text
+            zero_shot_request.prompt_text = args.prompt_text
+            prompt_speech = load_wav(args.prompt_wav, 16000)
+            zero_shot_request.prompt_audio = (prompt_speech.numpy() * (2**15)).astype(np.int16).tobytes()
+            request.zero_shot_request.CopyFrom(zero_shot_request)
+        elif args.mode == 'cross_lingual':
+            logging.info('send cross_lingual request')
+            cross_lingual_request = cosyvoice_pb2.crosslingualRequest()
+            cross_lingual_request.tts_text = args.tts_text
+            prompt_speech = load_wav(args.prompt_wav, 16000)
+            cross_lingual_request.prompt_audio = (prompt_speech.numpy() * (2**15)).astype(np.int16).tobytes()
+            request.cross_lingual_request.CopyFrom(cross_lingual_request)
+        else:
+            logging.info('send instruct request')
+            instruct_request = cosyvoice_pb2.instructRequest()
+            instruct_request.tts_text = args.tts_text
+            instruct_request.spk_id = args.spk_id
+            instruct_request.instruct_text = args.instruct_text
+            request.instruct_request.CopyFrom(instruct_request)
+
+        response = stub.Inference(request)
+        tts_audio = b''
+        for r in response:
+            tts_audio += r.tts_audio
+        tts_speech = torch.from_numpy(np.array(np.frombuffer(tts_audio, dtype=np.int16))).unsqueeze(dim=0)
+        logging.info('save response to {}'.format(args.tts_wav))
+        torchaudio.save(args.tts_wav, tts_speech, target_sr)
+        logging.info('get response')
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument('--host',
+                        type=str,
+                        default='0.0.0.0')
+    parser.add_argument('--port',
+                        type=int,
+                        default='50000')
+    parser.add_argument('--mode',
+                        default='sft',
+                        choices=['sft', 'zero_shot', 'cross_lingual', 'instruct'],
+                        help='request mode')
+    parser.add_argument('--tts_text',
+                        type=str,
+                        default='你好，我是通义千问语音合成大模型，请问有什么可以帮您的吗？')
+    parser.add_argument('--spk_id',
+                        type=str,
+                        default='中文女')
+    parser.add_argument('--prompt_text',
+                        type=str,
+                        default='希望你以后能够做的比我还好呦。')
+    parser.add_argument('--prompt_wav',
+                        type=str,
+                        default='../../../asset/zero_shot_prompt.wav')
+    parser.add_argument('--instruct_text',
+                        type=str,
+                        default='Theo \'Crimson\', is a fiery, passionate rebel leader. \
+                                 Fights with fervor for justice, but struggles with impulsiveness.')
+    parser.add_argument('--tts_wav',
+                        type=str,
+                        default='demo.wav')
+    args = parser.parse_args()
+    prompt_sr, target_sr = 16000, 22050
+    main()

third_party/Matcha-TTS/.env.example

@@ -0,0 +1,6 @@
+# example of file for storing private and user specific environment variables, like keys or system paths
+# rename it to ".env" (excluded from version control by default)
+# .env is loaded by train.py automatically
+# hydra allows you to reference variables in .yaml configs with special syntax: ${oc.env:MY_VAR}
+
+MY_VAR="/home/user/my/system/path"

third_party/Matcha-TTS/.github/PULL_REQUEST_TEMPLATE.md

@@ -0,0 +1,22 @@
+## What does this PR do?
+
+<!--
+Please include a summary of the change and which issue is fixed.
+Please also include relevant motivation and context.
+List any dependencies that are required for this change.
+List all the breaking changes introduced by this pull request.
+-->
+
+Fixes #\<issue_number>
+
+## Before submitting
+
+- [ ] Did you make sure **title is self-explanatory** and **the description concisely explains the PR**?
+- [ ] Did you make sure your **PR does only one thing**, instead of bundling different changes together?
+- [ ] Did you list all the **breaking changes** introduced by this pull request?
+- [ ] Did you **test your PR locally** with `pytest` command?
+- [ ] Did you **run pre-commit hooks** with `pre-commit run -a` command?
+
+## Did you have fun?
+
+Make sure you had fun coding 🙃

third_party/Matcha-TTS/.github/codecov.yml

@@ -0,0 +1,15 @@
+coverage:
+  status:
+    # measures overall project coverage
+    project:
+      default:
+        threshold: 100% # how much decrease in coverage is needed to not consider success
+
+    # measures PR or single commit coverage
+    patch:
+      default:
+        threshold: 100% # how much decrease in coverage is needed to not consider success
+
+
+    # project: off
+    # patch: off

third_party/Matcha-TTS/.github/dependabot.yml

@@ -0,0 +1,17 @@
+# To get started with Dependabot version updates, you'll need to specify which
+# package ecosystems to update and where the package manifests are located.
+# Please see the documentation for all configuration options:
+# https://docs.github.com/github/administering-a-repository/configuration-options-for-dependency-updates
+
+version: 2
+updates:
+  - package-ecosystem: "pip" # See documentation for possible values
+    directory: "/" # Location of package manifests
+    target-branch: "dev"
+    schedule:
+      interval: "daily"
+    ignore:
+      - dependency-name: "pytorch-lightning"
+        update-types: ["version-update:semver-patch"]
+      - dependency-name: "torchmetrics"
+        update-types: ["version-update:semver-patch"]

third_party/Matcha-TTS/.github/release-drafter.yml

@@ -0,0 +1,44 @@
+name-template: "v$RESOLVED_VERSION"
+tag-template: "v$RESOLVED_VERSION"
+
+categories:
+  - title: "🚀 Features"
+    labels:
+      - "feature"
+      - "enhancement"
+  - title: "🐛 Bug Fixes"
+    labels:
+      - "fix"
+      - "bugfix"
+      - "bug"
+  - title: "🧹 Maintenance"
+    labels:
+      - "maintenance"
+      - "dependencies"
+      - "refactoring"
+      - "cosmetic"
+      - "chore"
+  - title: "📝️ Documentation"
+    labels:
+      - "documentation"
+      - "docs"
+
+change-template: "- $TITLE @$AUTHOR (#$NUMBER)"
+change-title-escapes: '\<*_&' # You can add # and @ to disable mentions
+
+version-resolver:
+  major:
+    labels:
+      - "major"
+  minor:
+    labels:
+      - "minor"
+  patch:
+    labels:
+      - "patch"
+  default: patch
+
+template: |
+  ## Changes
+
+  $CHANGES

third_party/Matcha-TTS/.gitignore

@@ -0,0 +1,163 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+pip-wheel-metadata/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+.python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+### VisualStudioCode
+.vscode/*
+!.vscode/settings.json
+!.vscode/tasks.json
+!.vscode/launch.json
+!.vscode/extensions.json
+*.code-workspace
+**/.vscode
+
+# JetBrains
+.idea/
+
+# Data & Models
+*.h5
+*.tar
+*.tar.gz
+
+# Lightning-Hydra-Template
+configs/local/default.yaml
+/data/
+/logs/
+.env
+
+# Aim logging
+.aim
+
+# Cython complied files
+matcha/utils/monotonic_align/core.c
+
+# Ignoring hifigan checkpoint
+generator_v1
+g_02500000
+gradio_cached_examples/
+synth_output/

third_party/Matcha-TTS/.pre-commit-config.yaml

@@ -0,0 +1,59 @@
+default_language_version:
+  python: python3.10
+
+repos:
+  - repo: https://github.com/pre-commit/pre-commit-hooks
+    rev: v4.5.0
+    hooks:
+      # list of supported hooks: https://pre-commit.com/hooks.html
+      - id: trailing-whitespace
+      - id: end-of-file-fixer
+      # - id: check-docstring-first
+      - id: check-yaml
+      - id: debug-statements
+      - id: detect-private-key
+      - id: check-toml
+      - id: check-case-conflict
+      - id: check-added-large-files
+
+  # python code formatting
+  - repo: https://github.com/psf/black
+    rev: 23.12.1
+    hooks:
+      - id: black
+        args: [--line-length, "120"]
+
+  # python import sorting
+  - repo: https://github.com/PyCQA/isort
+    rev: 5.13.2
+    hooks:
+      - id: isort
+        args: ["--profile", "black", "--filter-files"]
+
+  # python upgrading syntax to newer version
+  - repo: https://github.com/asottile/pyupgrade
+    rev: v3.15.0
+    hooks:
+      - id: pyupgrade
+        args: [--py38-plus]
+
+  # python check (PEP8), programming errors and code complexity
+  - repo: https://github.com/PyCQA/flake8
+    rev: 7.0.0
+    hooks:
+      - id: flake8
+        args:
+          [
+            "--max-line-length", "120",
+            "--extend-ignore",
+            "E203,E402,E501,F401,F841,RST2,RST301",
+            "--exclude",
+            "logs/*,data/*,matcha/hifigan/*",
+          ]
+        additional_dependencies: [flake8-rst-docstrings==0.3.0]
+
+  # pylint
+  - repo: https://github.com/pycqa/pylint
+    rev: v3.0.3
+    hooks:
+    -   id: pylint

third_party/Matcha-TTS/.project-root

@@ -0,0 +1,2 @@
+# this file is required for inferring the project root directory
+# do not delete

third_party/Matcha-TTS/.pylintrc

@@ -0,0 +1,525 @@
+[MASTER]
+
+# A comma-separated list of package or module names from where C extensions may
+# be loaded. Extensions are loading into the active Python interpreter and may
+# run arbitrary code.
+extension-pkg-whitelist=
+
+# Add files or directories to the blacklist. They should be base names, not
+# paths.
+ignore=CVS
+
+# Add files or directories matching the regex patterns to the blacklist. The
+# regex matches against base names, not paths.
+ignore-patterns=
+
+# Python code to execute, usually for sys.path manipulation such as
+# pygtk.require().
+#init-hook=
+
+# Use multiple processes to speed up Pylint. Specifying 0 will auto-detect the
+# number of processors available to use.
+jobs=1
+
+# Control the amount of potential inferred values when inferring a single
+# object. This can help the performance when dealing with large functions or
+# complex, nested conditions.
+limit-inference-results=100
+
+# List of plugins (as comma separated values of python modules names) to load,
+# usually to register additional checkers.
+load-plugins=
+
+# Pickle collected data for later comparisons.
+persistent=yes
+
+# Specify a configuration file.
+#rcfile=
+
+# When enabled, pylint would attempt to guess common misconfiguration and emit
+# user-friendly hints instead of false-positive error messages.
+suggestion-mode=yes
+
+# Allow loading of arbitrary C extensions. Extensions are imported into the
+# active Python interpreter and may run arbitrary code.
+unsafe-load-any-extension=no
+
+
+[MESSAGES CONTROL]
+
+# Only show warnings with the listed confidence levels. Leave empty to show
+# all. Valid levels: HIGH, INFERENCE, INFERENCE_FAILURE, UNDEFINED.
+confidence=
+
+# Disable the message, report, category or checker with the given id(s). You
+# can either give multiple identifiers separated by comma (,) or put this
+# option multiple times (only on the command line, not in the configuration
+# file where it should appear only once). You can also use "--disable=all" to
+# disable everything first and then reenable specific checks. For example, if
+# you want to run only the similarities checker, you can use "--disable=all
+# --enable=similarities". If you want to run only the classes checker, but have
+# no Warning level messages displayed, use "--disable=all --enable=classes
+# --disable=W".
+disable=missing-docstring,
+        too-many-public-methods,
+        too-many-lines,
+        bare-except,
+        ## for avoiding weird p3.6 CI linter error
+        ## TODO: see later if we can remove this
+        assigning-non-slot,
+        unsupported-assignment-operation,
+        ## end
+        line-too-long,
+        fixme,
+        wrong-import-order,
+        ungrouped-imports,
+        wrong-import-position,
+        import-error,
+        invalid-name,
+        too-many-instance-attributes,
+        arguments-differ,
+        arguments-renamed,
+        no-name-in-module,
+        no-member,
+        unsubscriptable-object,
+        raw-checker-failed,
+        bad-inline-option,
+        locally-disabled,
+        file-ignored,
+        suppressed-message,
+        useless-suppression,
+        deprecated-pragma,
+        use-symbolic-message-instead,
+        useless-object-inheritance,
+        too-few-public-methods,
+        too-many-branches,
+        too-many-arguments,
+        too-many-locals,
+        too-many-statements,
+        duplicate-code,
+        not-callable,
+        import-outside-toplevel,
+        logging-fstring-interpolation,
+        logging-not-lazy,
+        unused-argument,
+        no-else-return,
+        chained-comparison,
+        redefined-outer-name
+
+# Enable the message, report, category or checker with the given id(s). You can
+# either give multiple identifier separated by comma (,) or put this option
+# multiple time (only on the command line, not in the configuration file where
+# it should appear only once). See also the "--disable" option for examples.
+enable=c-extension-no-member
+
+
+[REPORTS]
+
+# Python expression which should return a note less than 10 (10 is the highest
+# note). You have access to the variables errors warning, statement which
+# respectively contain the number of errors / warnings messages and the total
+# number of statements analyzed. This is used by the global evaluation report
+# (RP0004).
+evaluation=10.0 - ((float(5 * error + warning + refactor + convention) / statement) * 10)
+
+# Template used to display messages. This is a python new-style format string
+# used to format the message information. See doc for all details.
+#msg-template=
+
+# Set the output format. Available formats are text, parseable, colorized, json
+# and msvs (visual studio). You can also give a reporter class, e.g.
+# mypackage.mymodule.MyReporterClass.
+output-format=text
+
+# Tells whether to display a full report or only the messages.
+reports=no
+
+# Activate the evaluation score.
+score=yes
+
+
+[REFACTORING]
+
+# Maximum number of nested blocks for function / method body
+max-nested-blocks=5
+
+# Complete name of functions that never returns. When checking for
+# inconsistent-return-statements if a never returning function is called then
+# it will be considered as an explicit return statement and no message will be
+# printed.
+never-returning-functions=sys.exit
+
+
+[LOGGING]
+
+# Format style used to check logging format string. `old` means using %
+# formatting, while `new` is for `{}` formatting.
+logging-format-style=old
+
+# Logging modules to check that the string format arguments are in logging
+# function parameter format.
+logging-modules=logging
+
+
+[SPELLING]
+
+# Limits count of emitted suggestions for spelling mistakes.
+max-spelling-suggestions=4
+
+# Spelling dictionary name. Available dictionaries: none. To make it working
+# install python-enchant package..
+spelling-dict=
+
+# List of comma separated words that should not be checked.
+spelling-ignore-words=
+
+# A path to a file that contains private dictionary; one word per line.
+spelling-private-dict-file=
+
+# Tells whether to store unknown words to indicated private dictionary in
+# --spelling-private-dict-file option instead of raising a message.
+spelling-store-unknown-words=no
+
+
+[MISCELLANEOUS]
+
+# List of note tags to take in consideration, separated by a comma.
+notes=FIXME,
+      XXX,
+      TODO
+
+
+[TYPECHECK]
+
+# List of decorators that produce context managers, such as
+# contextlib.contextmanager. Add to this list to register other decorators that
+# produce valid context managers.
+contextmanager-decorators=contextlib.contextmanager
+
+# List of members which are set dynamically and missed by pylint inference
+# system, and so shouldn't trigger E1101 when accessed. Python regular
+# expressions are accepted.
+generated-members=numpy.*,torch.*
+
+# Tells whether missing members accessed in mixin class should be ignored. A
+# mixin class is detected if its name ends with "mixin" (case insensitive).
+ignore-mixin-members=yes
+
+# Tells whether to warn about missing members when the owner of the attribute
+# is inferred to be None.
+ignore-none=yes
+
+# This flag controls whether pylint should warn about no-member and similar
+# checks whenever an opaque object is returned when inferring. The inference
+# can return multiple potential results while evaluating a Python object, but
+# some branches might not be evaluated, which results in partial inference. In
+# that case, it might be useful to still emit no-member and other checks for
+# the rest of the inferred objects.
+ignore-on-opaque-inference=yes
+
+# List of class names for which member attributes should not be checked (useful
+# for classes with dynamically set attributes). This supports the use of
+# qualified names.
+ignored-classes=optparse.Values,thread._local,_thread._local
+
+# List of module names for which member attributes should not be checked
+# (useful for modules/projects where namespaces are manipulated during runtime
+# and thus existing member attributes cannot be deduced by static analysis. It
+# supports qualified module names, as well as Unix pattern matching.
+ignored-modules=
+
+# Show a hint with possible names when a member name was not found. The aspect
+# of finding the hint is based on edit distance.
+missing-member-hint=yes
+
+# The minimum edit distance a name should have in order to be considered a
+# similar match for a missing member name.
+missing-member-hint-distance=1
+
+# The total number of similar names that should be taken in consideration when
+# showing a hint for a missing member.
+missing-member-max-choices=1
+
+
+[VARIABLES]
+
+# List of additional names supposed to be defined in builtins. Remember that
+# you should avoid defining new builtins when possible.
+additional-builtins=
+
+# Tells whether unused global variables should be treated as a violation.
+allow-global-unused-variables=yes
+
+# List of strings which can identify a callback function by name. A callback
+# name must start or end with one of those strings.
+callbacks=cb_,
+          _cb
+
+# A regular expression matching the name of dummy variables (i.e. expected to
+# not be used).
+dummy-variables-rgx=_+$|(_[a-zA-Z0-9_]*[a-zA-Z0-9]+?$)|dummy|^ignored_|^unused_
+
+# Argument names that match this expression will be ignored. Default to name
+# with leading underscore.
+ignored-argument-names=_.*|^ignored_|^unused_
+
+# Tells whether we should check for unused import in __init__ files.
+init-import=no
+
+# List of qualified module names which can have objects that can redefine
+# builtins.
+redefining-builtins-modules=six.moves,past.builtins,future.builtins,builtins,io
+
+
+[FORMAT]
+
+# Expected format of line ending, e.g. empty (any line ending), LF or CRLF.
+expected-line-ending-format=
+
+# Regexp for a line that is allowed to be longer than the limit.
+ignore-long-lines=^\s*(# )?<?https?://\S+>?$
+
+# Number of spaces of indent required inside a hanging or continued line.
+indent-after-paren=4
+
+# String used as indentation unit. This is usually "    " (4 spaces) or "\t" (1
+# tab).
+indent-string='    '
+
+# Maximum number of characters on a single line.
+max-line-length=120
+
+# Maximum number of lines in a module.
+max-module-lines=1000
+
+# Allow the body of a class to be on the same line as the declaration if body
+# contains single statement.
+single-line-class-stmt=no
+
+# Allow the body of an if to be on the same line as the test if there is no
+# else.
+single-line-if-stmt=no
+
+
+[SIMILARITIES]
+
+# Ignore comments when computing similarities.
+ignore-comments=yes
+
+# Ignore docstrings when computing similarities.
+ignore-docstrings=yes
+
+# Ignore imports when computing similarities.
+ignore-imports=no
+
+# Minimum lines number of a similarity.
+min-similarity-lines=4
+
+
+[BASIC]
+
+# Naming style matching correct argument names.
+argument-naming-style=snake_case
+
+# Regular expression matching correct argument names. Overrides argument-
+# naming-style.
+argument-rgx=[a-z_][a-z0-9_]{0,30}$
+
+# Naming style matching correct attribute names.
+attr-naming-style=snake_case
+
+# Regular expression matching correct attribute names. Overrides attr-naming-
+# style.
+#attr-rgx=
+
+# Bad variable names which should always be refused, separated by a comma.
+bad-names=
+
+# Naming style matching correct class attribute names.
+class-attribute-naming-style=any
+
+# Regular expression matching correct class attribute names. Overrides class-
+# attribute-naming-style.
+#class-attribute-rgx=
+
+# Naming style matching correct class names.
+class-naming-style=PascalCase
+
+# Regular expression matching correct class names. Overrides class-naming-
+# style.
+#class-rgx=
+
+# Naming style matching correct constant names.
+const-naming-style=UPPER_CASE
+
+# Regular expression matching correct constant names. Overrides const-naming-
+# style.
+#const-rgx=
+
+# Minimum line length for functions/classes that require docstrings, shorter
+# ones are exempt.
+docstring-min-length=-1
+
+# Naming style matching correct function names.
+function-naming-style=snake_case
+
+# Regular expression matching correct function names. Overrides function-
+# naming-style.
+#function-rgx=
+
+# Good variable names which should always be accepted, separated by a comma.
+good-names=i,
+           j,
+           k,
+           x,
+           ex,
+           Run,
+           _
+
+# Include a hint for the correct naming format with invalid-name.
+include-naming-hint=no
+
+# Naming style matching correct inline iteration names.
+inlinevar-naming-style=any
+
+# Regular expression matching correct inline iteration names. Overrides
+# inlinevar-naming-style.
+#inlinevar-rgx=
+
+# Naming style matching correct method names.
+method-naming-style=snake_case
+
+# Regular expression matching correct method names. Overrides method-naming-
+# style.
+#method-rgx=
+
+# Naming style matching correct module names.
+module-naming-style=snake_case
+
+# Regular expression matching correct module names. Overrides module-naming-
+# style.
+#module-rgx=
+
+# Colon-delimited sets of names that determine each other's naming style when
+# the name regexes allow several styles.
+name-group=
+
+# Regular expression which should only match function or class names that do
+# not require a docstring.
+no-docstring-rgx=^_
+
+# List of decorators that produce properties, such as abc.abstractproperty. Add
+# to this list to register other decorators that produce valid properties.
+# These decorators are taken in consideration only for invalid-name.
+property-classes=abc.abstractproperty
+
+# Naming style matching correct variable names.
+variable-naming-style=snake_case
+
+# Regular expression matching correct variable names. Overrides variable-
+# naming-style.
+variable-rgx=[a-z_][a-z0-9_]{0,30}$
+
+
+[STRING]
+
+# This flag controls whether the implicit-str-concat-in-sequence should
+# generate a warning on implicit string concatenation in sequences defined over
+# several lines.
+check-str-concat-over-line-jumps=no
+
+
+[IMPORTS]
+
+# Allow wildcard imports from modules that define __all__.
+allow-wildcard-with-all=no
+
+# Analyse import fallback blocks. This can be used to support both Python 2 and
+# 3 compatible code, which means that the block might have code that exists
+# only in one or another interpreter, leading to false positives when analysed.
+analyse-fallback-blocks=no
+
+# Deprecated modules which should not be used, separated by a comma.
+deprecated-modules=optparse,tkinter.tix
+
+# Create a graph of external dependencies in the given file (report RP0402 must
+# not be disabled).
+ext-import-graph=
+
+# Create a graph of every (i.e. internal and external) dependencies in the
+# given file (report RP0402 must not be disabled).
+import-graph=
+
+# Create a graph of internal dependencies in the given file (report RP0402 must
+# not be disabled).
+int-import-graph=
+
+# Force import order to recognize a module as part of the standard
+# compatibility libraries.
+known-standard-library=
+
+# Force import order to recognize a module as part of a third party library.
+known-third-party=enchant
+
+
+[CLASSES]
+
+# List of method names used to declare (i.e. assign) instance attributes.
+defining-attr-methods=__init__,
+                      __new__,
+                      setUp
+
+# List of member names, which should be excluded from the protected access
+# warning.
+exclude-protected=_asdict,
+                  _fields,
+                  _replace,
+                  _source,
+                  _make
+
+# List of valid names for the first argument in a class method.
+valid-classmethod-first-arg=cls
+
+# List of valid names for the first argument in a metaclass class method.
+valid-metaclass-classmethod-first-arg=cls
+
+
+[DESIGN]
+
+# Maximum number of arguments for function / method.
+max-args=5
+
+# Maximum number of attributes for a class (see R0902).
+max-attributes=7
+
+# Maximum number of boolean expressions in an if statement.
+max-bool-expr=5
+
+# Maximum number of branch for function / method body.
+max-branches=12
+
+# Maximum number of locals for function / method body.
+max-locals=15
+
+# Maximum number of parents for a class (see R0901).
+max-parents=15
+
+# Maximum number of public methods for a class (see R0904).
+max-public-methods=20
+
+# Maximum number of return / yield for function / method body.
+max-returns=6
+
+# Maximum number of statements in function / method body.
+max-statements=50
+
+# Minimum number of public methods for a class (see R0903).
+min-public-methods=2
+
+
+[EXCEPTIONS]
+
+# Exceptions that will emit a warning when being caught. Defaults to
+# "BaseException, Exception".
+overgeneral-exceptions=builtins.BaseException,
+                       builtins.Exception

third_party/Matcha-TTS/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2023 Shivam Mehta
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

third_party/Matcha-TTS/MANIFEST.in

@@ -0,0 +1,14 @@
+include README.md
+include LICENSE.txt
+include requirements.*.txt
+include *.cff
+include requirements.txt
+include matcha/VERSION
+recursive-include matcha *.json
+recursive-include matcha *.html
+recursive-include matcha *.png
+recursive-include matcha *.md
+recursive-include matcha *.py
+recursive-include matcha *.pyx
+recursive-exclude tests *
+prune tests*

third_party/Matcha-TTS/Makefile

@@ -0,0 +1,42 @@
+
+help:  ## Show help
+	@grep -E '^[.a-zA-Z_-]+:.*?## .*$$' $(MAKEFILE_LIST) | awk 'BEGIN {FS = ":.*?## "}; {printf "\033[36m%-30s\033[0m %s\n", $$1, $$2}'
+
+clean: ## Clean autogenerated files
+	rm -rf dist
+	find . -type f -name "*.DS_Store" -ls -delete
+	find . | grep -E "(__pycache__|\.pyc|\.pyo)" | xargs rm -rf
+	find . | grep -E ".pytest_cache" | xargs rm -rf
+	find . | grep -E ".ipynb_checkpoints" | xargs rm -rf
+	rm -f .coverage
+
+clean-logs: ## Clean logs
+	rm -rf logs/**
+
+create-package: ## Create wheel and tar gz
+	rm -rf dist/
+	python setup.py bdist_wheel --plat-name=manylinux1_x86_64
+	python setup.py sdist
+	python -m twine upload  dist/* --verbose --skip-existing
+
+format: ## Run pre-commit hooks
+	pre-commit run -a
+
+sync: ## Merge changes from main branch to your current branch
+	git pull
+	git pull origin main
+
+test: ## Run not slow tests
+	pytest -k "not slow"
+
+test-full: ## Run all tests
+	pytest
+
+train-ljspeech: ## Train the model
+	python matcha/train.py experiment=ljspeech
+
+train-ljspeech-min: ## Train the model with minimum memory
+	python matcha/train.py experiment=ljspeech_min_memory
+
+start_app: ## Start the app
+	python matcha/app.py

third_party/Matcha-TTS/README.md

@@ -0,0 +1,278 @@
+<div align="center">
+
+# 🍵 Matcha-TTS: A fast TTS architecture with conditional flow matching
+
+### [Shivam Mehta](https://www.kth.se/profile/smehta), [Ruibo Tu](https://www.kth.se/profile/ruibo), [Jonas Beskow](https://www.kth.se/profile/beskow), [Éva Székely](https://www.kth.se/profile/szekely), and [Gustav Eje Henter](https://people.kth.se/~ghe/)
+
+[![python](https://img.shields.io/badge/-Python_3.10-blue?logo=python&logoColor=white)](https://www.python.org/downloads/release/python-3100/)
+[![pytorch](https://img.shields.io/badge/PyTorch_2.0+-ee4c2c?logo=pytorch&logoColor=white)](https://pytorch.org/get-started/locally/)
+[![lightning](https://img.shields.io/badge/-Lightning_2.0+-792ee5?logo=pytorchlightning&logoColor=white)](https://pytorchlightning.ai/)
+[![hydra](https://img.shields.io/badge/Config-Hydra_1.3-89b8cd)](https://hydra.cc/)
+[![black](https://img.shields.io/badge/Code%20Style-Black-black.svg?labelColor=gray)](https://black.readthedocs.io/en/stable/)
+[![isort](https://img.shields.io/badge/%20imports-isort-%231674b1?style=flat&labelColor=ef8336)](https://pycqa.github.io/isort/)
+
+<p style="text-align: center;">
+  <img src="https://shivammehta25.github.io/Matcha-TTS/images/logo.png" height="128"/>
+</p>
+
+</div>
+
+> This is the official code implementation of 🍵 Matcha-TTS [ICASSP 2024].
+
+We propose 🍵 Matcha-TTS, a new approach to non-autoregressive neural TTS, that uses [conditional flow matching](https://arxiv.org/abs/2210.02747) (similar to [rectified flows](https://arxiv.org/abs/2209.03003)) to speed up ODE-based speech synthesis. Our method:
+
+- Is probabilistic
+- Has compact memory footprint
+- Sounds highly natural
+- Is very fast to synthesise from
+
+Check out our [demo page](https://shivammehta25.github.io/Matcha-TTS) and read [our ICASSP 2024 paper](https://arxiv.org/abs/2309.03199) for more details.
+
+[Pre-trained models](https://drive.google.com/drive/folders/17C_gYgEHOxI5ZypcfE_k1piKCtyR0isJ?usp=sharing) will be automatically downloaded with the CLI or gradio interface.
+
+You can also [try 🍵 Matcha-TTS in your browser on HuggingFace 🤗 spaces](https://huggingface.co/spaces/shivammehta25/Matcha-TTS).
+
+## Teaser video
+
+[![Watch the video](https://img.youtube.com/vi/xmvJkz3bqw0/hqdefault.jpg)](https://youtu.be/xmvJkz3bqw0)
+
+## Installation
+
+1. Create an environment (suggested but optional)
+
+```
+conda create -n matcha-tts python=3.10 -y
+conda activate matcha-tts
+```
+
+2. Install Matcha TTS using pip or from source
+
+```bash
+pip install matcha-tts
+```
+
+from source
+
+```bash
+pip install git+https://github.com/shivammehta25/Matcha-TTS.git
+cd Matcha-TTS
+pip install -e .
+```
+
+3. Run CLI / gradio app / jupyter notebook
+
+```bash
+# This will download the required models
+matcha-tts --text "<INPUT TEXT>"
+```
+
+or
+
+```bash
+matcha-tts-app
+```
+
+or open `synthesis.ipynb` on jupyter notebook
+
+### CLI Arguments
+
+- To synthesise from given text, run:
+
+```bash
+matcha-tts --text "<INPUT TEXT>"
+```
+
+- To synthesise from a file, run:
+
+```bash
+matcha-tts --file <PATH TO FILE>
+```
+
+- To batch synthesise from a file, run:
+
+```bash
+matcha-tts --file <PATH TO FILE> --batched
+```
+
+Additional arguments
+
+- Speaking rate
+
+```bash
+matcha-tts --text "<INPUT TEXT>" --speaking_rate 1.0
+```
+
+- Sampling temperature
+
+```bash
+matcha-tts --text "<INPUT TEXT>" --temperature 0.667
+```
+
+- Euler ODE solver steps
+
+```bash
+matcha-tts --text "<INPUT TEXT>" --steps 10
+```
+
+## Train with your own dataset
+
+Let's assume we are training with LJ Speech
+
+1. Download the dataset from [here](https://keithito.com/LJ-Speech-Dataset/), extract it to `data/LJSpeech-1.1`, and prepare the file lists to point to the extracted data like for [item 5 in the setup of the NVIDIA Tacotron 2 repo](https://github.com/NVIDIA/tacotron2#setup).
+
+2. Clone and enter the Matcha-TTS repository
+
+```bash
+git clone https://github.com/shivammehta25/Matcha-TTS.git
+cd Matcha-TTS
+```
+
+3. Install the package from source
+
+```bash
+pip install -e .
+```
+
+4. Go to `configs/data/ljspeech.yaml` and change
+
+```yaml
+train_filelist_path: data/filelists/ljs_audio_text_train_filelist.txt
+valid_filelist_path: data/filelists/ljs_audio_text_val_filelist.txt
+```
+
+5. Generate normalisation statistics with the yaml file of dataset configuration
+
+```bash
+matcha-data-stats -i ljspeech.yaml
+# Output:
+#{'mel_mean': -5.53662231756592, 'mel_std': 2.1161014277038574}
+```
+
+Update these values in `configs/data/ljspeech.yaml` under `data_statistics` key.
+
+```bash
+data_statistics:  # Computed for ljspeech dataset
+  mel_mean: -5.536622
+  mel_std: 2.116101
+```
+
+to the paths of your train and validation filelists.
+
+6. Run the training script
+
+```bash
+make train-ljspeech
+```
+
+or
+
+```bash
+python matcha/train.py experiment=ljspeech
+```
+
+- for a minimum memory run
+
+```bash
+python matcha/train.py experiment=ljspeech_min_memory
+```
+
+- for multi-gpu training, run
+
+```bash
+python matcha/train.py experiment=ljspeech trainer.devices=[0,1]
+```
+
+7. Synthesise from the custom trained model
+
+```bash
+matcha-tts --text "<INPUT TEXT>" --checkpoint_path <PATH TO CHECKPOINT>
+```
+
+## ONNX support
+
+> Special thanks to [@mush42](https://github.com/mush42) for implementing ONNX export and inference support.
+
+It is possible to export Matcha checkpoints to [ONNX](https://onnx.ai/), and run inference on the exported ONNX graph.
+
+### ONNX export
+
+To export a checkpoint to ONNX, first install ONNX with
+
+```bash
+pip install onnx
+```
+
+then run the following:
+
+```bash
+python3 -m matcha.onnx.export matcha.ckpt model.onnx --n-timesteps 5
+```
+
+Optionally, the ONNX exporter accepts **vocoder-name** and **vocoder-checkpoint** arguments. This enables you to embed the vocoder in the exported graph and generate waveforms in a single run (similar to end-to-end TTS systems).
+
+**Note** that `n_timesteps` is treated as a hyper-parameter rather than a model input. This means you should specify it during export (not during inference). If not specified, `n_timesteps` is set to **5**.
+
+**Important**: for now, torch>=2.1.0 is needed for export since the `scaled_product_attention` operator is not exportable in older versions. Until the final version is released, those who want to export their models must install torch>=2.1.0 manually as a pre-release.
+
+### ONNX Inference
+
+To run inference on the exported model, first install `onnxruntime` using
+
+```bash
+pip install onnxruntime
+pip install onnxruntime-gpu  # for GPU inference
+```
+
+then use the following:
+
+```bash
+python3 -m matcha.onnx.infer model.onnx --text "hey" --output-dir ./outputs
+```
+
+You can also control synthesis parameters:
+
+```bash
+python3 -m matcha.onnx.infer model.onnx --text "hey" --output-dir ./outputs --temperature 0.4 --speaking_rate 0.9 --spk 0
+```
+
+To run inference on **GPU**, make sure to install **onnxruntime-gpu** package, and then pass `--gpu` to the inference command:
+
+```bash
+python3 -m matcha.onnx.infer model.onnx --text "hey" --output-dir ./outputs --gpu
+```
+
+If you exported only Matcha to ONNX, this will write mel-spectrogram as graphs and `numpy` arrays to the output directory.
+If you embedded the vocoder in the exported graph, this will write `.wav` audio files to the output directory.
+
+If you exported only Matcha to ONNX, and you want to run a full TTS pipeline, you can pass a path to a vocoder model in `ONNX` format:
+
+```bash
+python3 -m matcha.onnx.infer model.onnx --text "hey" --output-dir ./outputs --vocoder hifigan.small.onnx
+```
+
+This will write `.wav` audio files to the output directory.
+
+## Citation information
+
+If you use our code or otherwise find this work useful, please cite our paper:
+
+```text
+@inproceedings{mehta2024matcha,
+  title={Matcha-{TTS}: A fast {TTS} architecture with conditional flow matching},
+  author={Mehta, Shivam and Tu, Ruibo and Beskow, Jonas and Sz{\'e}kely, {\'E}va and Henter, Gustav Eje},
+  booktitle={Proc. ICASSP},
+  year={2024}
+}
+```
+
+## Acknowledgements
+
+Since this code uses [Lightning-Hydra-Template](https://github.com/ashleve/lightning-hydra-template), you have all the powers that come with it.
+
+Other source code we would like to acknowledge:
+
+- [Coqui-TTS](https://github.com/coqui-ai/TTS/tree/dev): For helping me figure out how to make cython binaries pip installable and encouragement
+- [Hugging Face Diffusers](https://huggingface.co/): For their awesome diffusers library and its components
+- [Grad-TTS](https://github.com/huawei-noah/Speech-Backbones/tree/main/Grad-TTS): For the monotonic alignment search source code
+- [torchdyn](https://github.com/DiffEqML/torchdyn): Useful for trying other ODE solvers during research and development
+- [labml.ai](https://nn.labml.ai/transformers/rope/index.html): For the RoPE implementation

third_party/Matcha-TTS/configs/__init__.py

@@ -0,0 +1 @@
+# this file is needed here to include configs when building project as a package

third_party/Matcha-TTS/configs/callbacks/default.yaml

@@ -0,0 +1,5 @@
+defaults:
+  - model_checkpoint.yaml
+  - model_summary.yaml
+  - rich_progress_bar.yaml
+  - _self_

third_party/Matcha-TTS/configs/callbacks/rich_progress_bar.yaml

@@ -0,0 +1,4 @@
+# https://lightning.ai/docs/pytorch/latest/api/lightning.pytorch.callbacks.RichProgressBar.html
+
+rich_progress_bar:
+  _target_: lightning.pytorch.callbacks.RichProgressBar

third_party/Matcha-TTS/configs/data/hi-fi_en-US_female.yaml

@@ -0,0 +1,14 @@
+defaults:
+  - ljspeech
+  - _self_
+
+# Dataset URL: https://ast-astrec.nict.go.jp/en/release/hi-fi-captain/
+_target_: matcha.data.text_mel_datamodule.TextMelDataModule
+name: hi-fi_en-US_female
+train_filelist_path: data/filelists/hi-fi-captain-en-us-female_train.txt
+valid_filelist_path: data/filelists/hi-fi-captain-en-us-female_val.txt
+batch_size: 32
+cleaners: [english_cleaners_piper]
+data_statistics:  # Computed for this dataset
+  mel_mean: -6.38385
+  mel_std: 2.541796

third_party/Matcha-TTS/configs/data/ljspeech.yaml

@@ -0,0 +1,21 @@
+_target_: matcha.data.text_mel_datamodule.TextMelDataModule
+name: ljspeech
+train_filelist_path: data/filelists/ljs_audio_text_train_filelist.txt
+valid_filelist_path: data/filelists/ljs_audio_text_val_filelist.txt
+batch_size: 32
+num_workers: 20
+pin_memory: True
+cleaners: [english_cleaners2]
+add_blank: True
+n_spks: 1
+n_fft: 1024
+n_feats: 80
+sample_rate: 22050
+hop_length: 256
+win_length: 1024
+f_min: 0
+f_max: 8000
+data_statistics:  # Computed for ljspeech dataset
+  mel_mean: -5.536622
+  mel_std: 2.116101
+seed: ${seed}

third_party/Matcha-TTS/configs/data/vctk.yaml

@@ -0,0 +1,14 @@
+defaults:
+  - ljspeech
+  - _self_
+
+_target_: matcha.data.text_mel_datamodule.TextMelDataModule
+name: vctk
+train_filelist_path: data/filelists/vctk_audio_sid_text_train_filelist.txt
+valid_filelist_path: data/filelists/vctk_audio_sid_text_val_filelist.txt
+batch_size: 32
+add_blank: True
+n_spks: 109
+data_statistics:  # Computed for vctk dataset
+  mel_mean: -6.630575
+  mel_std: 2.482914

third_party/Matcha-TTS/configs/debug/default.yaml

@@ -0,0 +1,35 @@
+# @package _global_
+
+# default debugging setup, runs 1 full epoch
+# other debugging configs can inherit from this one
+
+# overwrite task name so debugging logs are stored in separate folder
+task_name: "debug"
+
+# disable callbacks and loggers during debugging
+# callbacks: null
+# logger: null
+
+extras:
+  ignore_warnings: False
+  enforce_tags: False
+
+# sets level of all command line loggers to 'DEBUG'
+# https://hydra.cc/docs/tutorials/basic/running_your_app/logging/
+hydra:
+  job_logging:
+    root:
+      level: DEBUG
+
+  # use this to also set hydra loggers to 'DEBUG'
+  # verbose: True
+
+trainer:
+  max_epochs: 1
+  accelerator: cpu # debuggers don't like gpus
+  devices: 1 # debuggers don't like multiprocessing
+  detect_anomaly: true # raise exception if NaN or +/-inf is detected in any tensor
+
+data:
+  num_workers: 0 # debuggers don't like multiprocessing
+  pin_memory: False # disable gpu memory pin

third_party/Matcha-TTS/configs/debug/fdr.yaml

@@ -0,0 +1,9 @@
+# @package _global_
+
+# runs 1 train, 1 validation and 1 test step
+
+defaults:
+  - default
+
+trainer:
+  fast_dev_run: true

third_party/Matcha-TTS/configs/debug/limit.yaml

@@ -0,0 +1,12 @@
+# @package _global_
+
+# uses only 1% of the training data and 5% of validation/test data
+
+defaults:
+  - default
+
+trainer:
+  max_epochs: 3
+  limit_train_batches: 0.01
+  limit_val_batches: 0.05
+  limit_test_batches: 0.05

third_party/Matcha-TTS/configs/debug/overfit.yaml

@@ -0,0 +1,13 @@
+# @package _global_
+
+# overfits to 3 batches
+
+defaults:
+  - default
+
+trainer:
+  max_epochs: 20
+  overfit_batches: 3
+
+# model ckpt and early stopping need to be disabled during overfitting
+callbacks: null

third_party/Matcha-TTS/configs/debug/profiler.yaml

@@ -0,0 +1,15 @@
+# @package _global_
+
+# runs with execution time profiling
+
+defaults:
+  - default
+
+trainer:
+  max_epochs: 1
+  # profiler: "simple"
+  profiler: "advanced"
+  # profiler: "pytorch"
+  accelerator: gpu
+
+  limit_train_batches: 0.02

third_party/Matcha-TTS/configs/eval.yaml

@@ -0,0 +1,18 @@
+# @package _global_
+
+defaults:
+  - _self_
+  - data: mnist # choose datamodule with `test_dataloader()` for evaluation
+  - model: mnist
+  - logger: null
+  - trainer: default
+  - paths: default
+  - extras: default
+  - hydra: default
+
+task_name: "eval"
+
+tags: ["dev"]
+
+# passing checkpoint path is necessary for evaluation
+ckpt_path: ???

third_party/Matcha-TTS/configs/experiment/hifi_dataset_piper_phonemizer.yaml

@@ -0,0 +1,14 @@
+# @package _global_
+
+# to execute this experiment run:
+# python train.py experiment=multispeaker
+
+defaults:
+  - override /data: hi-fi_en-US_female.yaml
+
+# all parameters below will be merged with parameters from default configurations set above
+# this allows you to overwrite only specified parameters
+
+tags: ["hi-fi", "single_speaker", "piper_phonemizer", "en_US", "female"]
+
+run_name: hi-fi_en-US_female_piper_phonemizer

third_party/Matcha-TTS/configs/experiment/ljspeech.yaml

@@ -0,0 +1,14 @@
+# @package _global_
+
+# to execute this experiment run:
+# python train.py experiment=multispeaker
+
+defaults:
+  - override /data: ljspeech.yaml
+
+# all parameters below will be merged with parameters from default configurations set above
+# this allows you to overwrite only specified parameters
+
+tags: ["ljspeech"]
+
+run_name: ljspeech

third_party/Matcha-TTS/configs/experiment/ljspeech_min_memory.yaml

@@ -0,0 +1,18 @@
+# @package _global_
+
+# to execute this experiment run:
+# python train.py experiment=multispeaker
+
+defaults:
+  - override /data: ljspeech.yaml
+
+# all parameters below will be merged with parameters from default configurations set above
+# this allows you to overwrite only specified parameters
+
+tags: ["ljspeech"]
+
+run_name: ljspeech_min
+
+
+model:
+  out_size: 172

third_party/Matcha-TTS/configs/experiment/multispeaker.yaml

@@ -0,0 +1,14 @@
+# @package _global_
+
+# to execute this experiment run:
+# python train.py experiment=multispeaker
+
+defaults:
+  - override /data: vctk.yaml
+
+# all parameters below will be merged with parameters from default configurations set above
+# this allows you to overwrite only specified parameters
+
+tags: ["multispeaker"]
+
+run_name: multispeaker

third_party/Matcha-TTS/configs/extras/default.yaml

@@ -0,0 +1,8 @@
+# disable python warnings if they annoy you
+ignore_warnings: False
+
+# ask user for tags if none are provided in the config
+enforce_tags: True
+
+# pretty print config tree at the start of the run using Rich library
+print_config: True

third_party/Matcha-TTS/configs/hparams_search/mnist_optuna.yaml

@@ -0,0 +1,52 @@
+# @package _global_
+
+# example hyperparameter optimization of some experiment with Optuna:
+# python train.py -m hparams_search=mnist_optuna experiment=example
+
+defaults:
+  - override /hydra/sweeper: optuna
+
+# choose metric which will be optimized by Optuna
+# make sure this is the correct name of some metric logged in lightning module!
+optimized_metric: "val/acc_best"
+
+# here we define Optuna hyperparameter search
+# it optimizes for value returned from function with @hydra.main decorator
+# docs: https://hydra.cc/docs/next/plugins/optuna_sweeper
+hydra:
+  mode: "MULTIRUN" # set hydra to multirun by default if this config is attached
+
+  sweeper:
+    _target_: hydra_plugins.hydra_optuna_sweeper.optuna_sweeper.OptunaSweeper
+
+    # storage URL to persist optimization results
+    # for example, you can use SQLite if you set 'sqlite:///example.db'
+    storage: null
+
+    # name of the study to persist optimization results
+    study_name: null
+
+    # number of parallel workers
+    n_jobs: 1
+
+    # 'minimize' or 'maximize' the objective
+    direction: maximize
+
+    # total number of runs that will be executed
+    n_trials: 20
+
+    # choose Optuna hyperparameter sampler
+    # you can choose bayesian sampler (tpe), random search (without optimization), grid sampler, and others
+    # docs: https://optuna.readthedocs.io/en/stable/reference/samplers.html
+    sampler:
+      _target_: optuna.samplers.TPESampler
+      seed: 1234
+      n_startup_trials: 10 # number of random sampling runs before optimization starts
+
+    # define hyperparameter search space
+    params:
+      model.optimizer.lr: interval(0.0001, 0.1)
+      data.batch_size: choice(32, 64, 128, 256)
+      model.net.lin1_size: choice(64, 128, 256)
+      model.net.lin2_size: choice(64, 128, 256)
+      model.net.lin3_size: choice(32, 64, 128, 256)

third_party/Matcha-TTS/configs/hydra/default.yaml

@@ -0,0 +1,19 @@
+# https://hydra.cc/docs/configure_hydra/intro/
+
+# enable color logging
+defaults:
+  - override hydra_logging: colorlog
+  - override job_logging: colorlog
+
+# output directory, generated dynamically on each run
+run:
+  dir: ${paths.log_dir}/${task_name}/${run_name}/runs/${now:%Y-%m-%d}_${now:%H-%M-%S}
+sweep:
+  dir: ${paths.log_dir}/${task_name}/${run_name}/multiruns/${now:%Y-%m-%d}_${now:%H-%M-%S}
+  subdir: ${hydra.job.num}
+
+job_logging:
+  handlers:
+    file:
+      # Incorporates fix from https://github.com/facebookresearch/hydra/pull/2242
+      filename: ${hydra.runtime.output_dir}/${hydra.job.name}.log

third_party/Matcha-TTS/configs/logger/aim.yaml

@@ -0,0 +1,28 @@
+# https://aimstack.io/
+
+# example usage in lightning module:
+# https://github.com/aimhubio/aim/blob/main/examples/pytorch_lightning_track.py
+
+# open the Aim UI with the following command (run in the folder containing the `.aim` folder):
+# `aim up`
+
+aim:
+  _target_: aim.pytorch_lightning.AimLogger
+  repo: ${paths.root_dir} # .aim folder will be created here
+  # repo: "aim://ip_address:port" # can instead provide IP address pointing to Aim remote tracking server which manages the repo, see https://aimstack.readthedocs.io/en/latest/using/remote_tracking.html#
+
+  # aim allows to group runs under experiment name
+  experiment: null # any string, set to "default" if not specified
+
+  train_metric_prefix: "train/"
+  val_metric_prefix: "val/"
+  test_metric_prefix: "test/"
+
+  # sets the tracking interval in seconds for system usage metrics (CPU, GPU, memory, etc.)
+  system_tracking_interval: 10 # set to null to disable system metrics tracking
+
+  # enable/disable logging of system params such as installed packages, git info, env vars, etc.
+  log_system_params: true
+
+  # enable/disable tracking console logs (default value is true)
+  capture_terminal_logs: false # set to false to avoid infinite console log loop issue https://github.com/aimhubio/aim/issues/2550

third_party/Matcha-TTS/matcha/hifigan/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2020 Jungil Kong
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

third_party/Matcha-TTS/matcha/hifigan/README.md

@@ -0,0 +1,101 @@
+# HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis
+
+### Jungil Kong, Jaehyeon Kim, Jaekyoung Bae
+
+In our [paper](https://arxiv.org/abs/2010.05646),
+we proposed HiFi-GAN: a GAN-based model capable of generating high fidelity speech efficiently.<br/>
+We provide our implementation and pretrained models as open source in this repository.
+
+**Abstract :**
+Several recent work on speech synthesis have employed generative adversarial networks (GANs) to produce raw waveforms.
+Although such methods improve the sampling efficiency and memory usage,
+their sample quality has not yet reached that of autoregressive and flow-based generative models.
+In this work, we propose HiFi-GAN, which achieves both efficient and high-fidelity speech synthesis.
+As speech audio consists of sinusoidal signals with various periods,
+we demonstrate that modeling periodic patterns of an audio is crucial for enhancing sample quality.
+A subjective human evaluation (mean opinion score, MOS) of a single speaker dataset indicates that our proposed method
+demonstrates similarity to human quality while generating 22.05 kHz high-fidelity audio 167.9 times faster than
+real-time on a single V100 GPU. We further show the generality of HiFi-GAN to the mel-spectrogram inversion of unseen
+speakers and end-to-end speech synthesis. Finally, a small footprint version of HiFi-GAN generates samples 13.4 times
+faster than real-time on CPU with comparable quality to an autoregressive counterpart.
+
+Visit our [demo website](https://jik876.github.io/hifi-gan-demo/) for audio samples.
+
+## Pre-requisites
+
+1. Python >= 3.6
+2. Clone this repository.
+3. Install python requirements. Please refer [requirements.txt](requirements.txt)
+4. Download and extract the [LJ Speech dataset](https://keithito.com/LJ-Speech-Dataset/).
+   And move all wav files to `LJSpeech-1.1/wavs`
+
+## Training
+
+```
+python train.py --config config_v1.json
+```
+
+To train V2 or V3 Generator, replace `config_v1.json` with `config_v2.json` or `config_v3.json`.<br>
+Checkpoints and copy of the configuration file are saved in `cp_hifigan` directory by default.<br>
+You can change the path by adding `--checkpoint_path` option.
+
+Validation loss during training with V1 generator.<br>
+![validation loss](./validation_loss.png)
+
+## Pretrained Model
+
+You can also use pretrained models we provide.<br/>
+[Download pretrained models](https://drive.google.com/drive/folders/1-eEYTB5Av9jNql0WGBlRoi-WH2J7bp5Y?usp=sharing)<br/>
+Details of each folder are as in follows:
+
+| Folder Name  | Generator | Dataset   | Fine-Tuned                                             |
+| ------------ | --------- | --------- | ------------------------------------------------------ |
+| LJ_V1        | V1        | LJSpeech  | No                                                     |
+| LJ_V2        | V2        | LJSpeech  | No                                                     |
+| LJ_V3        | V3        | LJSpeech  | No                                                     |
+| LJ_FT_T2_V1  | V1        | LJSpeech  | Yes ([Tacotron2](https://github.com/NVIDIA/tacotron2)) |
+| LJ_FT_T2_V2  | V2        | LJSpeech  | Yes ([Tacotron2](https://github.com/NVIDIA/tacotron2)) |
+| LJ_FT_T2_V3  | V3        | LJSpeech  | Yes ([Tacotron2](https://github.com/NVIDIA/tacotron2)) |
+| VCTK_V1      | V1        | VCTK      | No                                                     |
+| VCTK_V2      | V2        | VCTK      | No                                                     |
+| VCTK_V3      | V3        | VCTK      | No                                                     |
+| UNIVERSAL_V1 | V1        | Universal | No                                                     |
+
+We provide the universal model with discriminator weights that can be used as a base for transfer learning to other datasets.
+
+## Fine-Tuning
+
+1. Generate mel-spectrograms in numpy format using [Tacotron2](https://github.com/NVIDIA/tacotron2) with teacher-forcing.<br/>
+   The file name of the generated mel-spectrogram should match the audio file and the extension should be `.npy`.<br/>
+   Example:
+   `   Audio File : LJ001-0001.wav
+Mel-Spectrogram File : LJ001-0001.npy`
+2. Create `ft_dataset` folder and copy the generated mel-spectrogram files into it.<br/>
+3. Run the following command.
+   ```
+   python train.py --fine_tuning True --config config_v1.json
+   ```
+   For other command line options, please refer to the training section.
+
+## Inference from wav file
+
+1. Make `test_files` directory and copy wav files into the directory.
+2. Run the following command.
+   `   python inference.py --checkpoint_file [generator checkpoint file path]`
+   Generated wav files are saved in `generated_files` by default.<br>
+   You can change the path by adding `--output_dir` option.
+
+## Inference for end-to-end speech synthesis
+
+1. Make `test_mel_files` directory and copy generated mel-spectrogram files into the directory.<br>
+   You can generate mel-spectrograms using [Tacotron2](https://github.com/NVIDIA/tacotron2),
+   [Glow-TTS](https://github.com/jaywalnut310/glow-tts) and so forth.
+2. Run the following command.
+   `   python inference_e2e.py --checkpoint_file [generator checkpoint file path]`
+   Generated wav files are saved in `generated_files_from_mel` by default.<br>
+   You can change the path by adding `--output_dir` option.
+
+## Acknowledgements
+
+We referred to [WaveGlow](https://github.com/NVIDIA/waveglow), [MelGAN](https://github.com/descriptinc/melgan-neurips)
+and [Tacotron2](https://github.com/NVIDIA/tacotron2) to implement this.