Spaces:
Running
on
Zero
Running
on
Zero
Den Pavloff
commited on
Commit
·
a9ff153
1
Parent(s):
e4e9267
release Generation 3 models
Browse files- README.md +9 -1
- examples.yaml +86 -48
- model_config.yaml +52 -5
README.md
CHANGED
|
@@ -11,8 +11,16 @@ license: apache-2.0
|
|
| 11 |
models:
|
| 12 |
- nineninesix/kani-tts-370m
|
| 13 |
- nineninesix/kani-tts-450m-0.2-pt
|
| 14 |
-
- nineninesix/kani-tts-
|
| 15 |
- nvidia/nemo-nano-codec-22khz-0.6kbps-12.5fps
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
---
|
| 17 |
|
| 18 |
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
|
|
|
| 11 |
models:
|
| 12 |
- nineninesix/kani-tts-370m
|
| 13 |
- nineninesix/kani-tts-450m-0.2-pt
|
| 14 |
+
- nineninesix/kani-tts-400m-0.3-pt
|
| 15 |
- nvidia/nemo-nano-codec-22khz-0.6kbps-12.5fps
|
| 16 |
+
- nineninesix/kani-tts-400m-en
|
| 17 |
+
- nineninesix/kani-tts-400m-zh
|
| 18 |
+
- nineninesix/kani-tts-400m-de
|
| 19 |
+
- nineninesix/kani-tts-400m-es
|
| 20 |
+
- nineninesix/kani-tts-370m-expo2025-osaka-ja
|
| 21 |
+
- nineninesix/kani-tts-400m-ar
|
| 22 |
+
- nineninesix/kani-tts-400m-ko
|
| 23 |
+
|
| 24 |
---
|
| 25 |
|
| 26 |
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
examples.yaml
CHANGED
|
@@ -1,90 +1,128 @@
|
|
| 1 |
examples:
|
| 2 |
- text: >-
|
| 3 |
-
|
| 4 |
-
speaker_id: "
|
| 5 |
-
model: "KaniTTS"
|
| 6 |
-
temperature:
|
| 7 |
top_p: 0.95
|
| 8 |
repetition_penalty: 1.1
|
| 9 |
-
max_len:
|
| 10 |
|
| 11 |
- text: >-
|
| 12 |
-
|
| 13 |
-
speaker_id: "
|
| 14 |
-
model: "KaniTTS"
|
| 15 |
-
temperature:
|
| 16 |
top_p: 0.95
|
| 17 |
repetition_penalty: 1.1
|
| 18 |
-
max_len:
|
| 19 |
|
| 20 |
- text: >-
|
| 21 |
Holy fu* Oh my God! Don't you understand how dangerous it is, huh?
|
| 22 |
-
speaker_id: "Andrew
|
| 23 |
-
model: "KaniTTS"
|
| 24 |
-
temperature:
|
| 25 |
top_p: 0.95
|
| 26 |
repetition_penalty: 1.1
|
| 27 |
-
max_len:
|
| 28 |
|
| 29 |
- text: >-
|
| 30 |
-
|
| 31 |
-
speaker_id: "
|
| 32 |
-
model: "KaniTTS"
|
| 33 |
-
temperature:
|
| 34 |
top_p: 0.95
|
| 35 |
repetition_penalty: 1.1
|
| 36 |
-
max_len:
|
| 37 |
|
| 38 |
- text: >-
|
| 39 |
-
|
| 40 |
-
speaker_id: "
|
| 41 |
-
model: "KaniTTS"
|
| 42 |
-
temperature:
|
| 43 |
top_p: 0.95
|
| 44 |
repetition_penalty: 1.1
|
| 45 |
-
max_len:
|
| 46 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 47 |
|
| 48 |
- text: >-
|
| 49 |
-
|
| 50 |
-
speaker_id:
|
| 51 |
-
model: "KaniTTS"
|
| 52 |
-
temperature:
|
| 53 |
top_p: 0.95
|
| 54 |
repetition_penalty: 1.1
|
| 55 |
-
max_len:
|
|
|
|
| 56 |
|
| 57 |
- text: >-
|
| 58 |
-
|
| 59 |
-
speaker_id: "
|
| 60 |
-
model: "KaniTTS"
|
| 61 |
-
temperature:
|
| 62 |
top_p: 0.95
|
| 63 |
repetition_penalty: 1.1
|
| 64 |
-
max_len:
|
| 65 |
|
| 66 |
- text: >-
|
| 67 |
-
|
| 68 |
-
speaker_id: "
|
| 69 |
-
model: "KaniTTS"
|
| 70 |
-
temperature:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 71 |
top_p: 0.95
|
| 72 |
repetition_penalty: 1.1
|
| 73 |
-
max_len:
|
| 74 |
|
| 75 |
- text: >-
|
| 76 |
-
|
| 77 |
-
speaker_id: "
|
| 78 |
-
model: "KaniTTS"
|
| 79 |
-
temperature:
|
| 80 |
top_p: 0.95
|
| 81 |
repetition_penalty: 1.1
|
| 82 |
-
max_len:
|
| 83 |
|
| 84 |
- text: >-
|
| 85 |
-
|
| 86 |
-
|
| 87 |
-
|
|
|
|
| 88 |
top_p: 0.95
|
| 89 |
repetition_penalty: 1.1
|
| 90 |
-
max_len:
|
|
|
|
| 1 |
examples:
|
| 2 |
- text: >-
|
| 3 |
+
No, that does not make you a failure. No, sweetie, no. It just, uh, it just means that you're having a tough time...
|
| 4 |
+
speaker_id: "Andrew"
|
| 5 |
+
model: "KaniTTS English"
|
| 6 |
+
temperature: 0.6
|
| 7 |
top_p: 0.95
|
| 8 |
repetition_penalty: 1.1
|
| 9 |
+
max_len: 900
|
| 10 |
|
| 11 |
- text: >-
|
| 12 |
+
Anyway, um, so, um, tell me, tell me all about her. I mean, what's she like? Is she really, you know, pretty?
|
| 13 |
+
speaker_id: "Katie"
|
| 14 |
+
model: "KaniTTS English"
|
| 15 |
+
temperature: 0.6
|
| 16 |
top_p: 0.95
|
| 17 |
repetition_penalty: 1.1
|
| 18 |
+
max_len: 900
|
| 19 |
|
| 20 |
- text: >-
|
| 21 |
Holy fu* Oh my God! Don't you understand how dangerous it is, huh?
|
| 22 |
+
speaker_id: "Andrew"
|
| 23 |
+
model: "KaniTTS English"
|
| 24 |
+
temperature: 0.6
|
| 25 |
top_p: 0.95
|
| 26 |
repetition_penalty: 1.1
|
| 27 |
+
max_len: 900
|
| 28 |
|
| 29 |
- text: >-
|
| 30 |
+
有一眼阳光晒到脸上,暖洋洋个,好舒服呃。
|
| 31 |
+
speaker_id: "Ming (Mandarin)"
|
| 32 |
+
model: "KaniTTS Chinese"
|
| 33 |
+
temperature: 0.6
|
| 34 |
top_p: 0.95
|
| 35 |
repetition_penalty: 1.1
|
| 36 |
+
max_len: 800
|
| 37 |
|
| 38 |
- text: >-
|
| 39 |
+
老頭子雖然年紀大,但係仍然日日早起去老坑嘅菜園到做嘢。
|
| 40 |
+
speaker_id: "Mei (Cantonese)"
|
| 41 |
+
model: "KaniTTS Chinese"
|
| 42 |
+
temperature: 0.6
|
| 43 |
top_p: 0.95
|
| 44 |
repetition_penalty: 1.1
|
| 45 |
+
max_len: 800
|
| 46 |
|
| 47 |
+
- text: >-
|
| 48 |
+
زقزقت عصافيرٌ مرِحة هذا الصباح على شجرة البلوط العتيقة خارج نافذتي.
|
| 49 |
+
speaker_id: null
|
| 50 |
+
model: "KaniTTS Arabic"
|
| 51 |
+
temperature: 0.6
|
| 52 |
+
top_p: 0.95
|
| 53 |
+
repetition_penalty: 1.1
|
| 54 |
+
max_len: 1000
|
| 55 |
|
| 56 |
- text: >-
|
| 57 |
+
مرحباً، اسمي تارا، وأنا نموذج لتوليد الصوت يمكنه أن يتحدث كالبشر تماماً.
|
| 58 |
+
speaker_id: null
|
| 59 |
+
model: "KaniTTS Arabic"
|
| 60 |
+
temperature: 0.6
|
| 61 |
top_p: 0.95
|
| 62 |
repetition_penalty: 1.1
|
| 63 |
+
max_len: 1000
|
| 64 |
+
|
| 65 |
|
| 66 |
- text: >-
|
| 67 |
+
Hast du jemals das Gefühl, dass die Zeit einfach davonrennt?
|
| 68 |
+
speaker_id: "Bert"
|
| 69 |
+
model: "KaniTTS Deutsch"
|
| 70 |
+
temperature: 0.6
|
| 71 |
top_p: 0.95
|
| 72 |
repetition_penalty: 1.1
|
| 73 |
+
max_len: 900
|
| 74 |
|
| 75 |
- text: >-
|
| 76 |
+
Was für ein unglaublicher Tag, voller kleiner Wunder!
|
| 77 |
+
speaker_id: "Thorsten (Hessisch)"
|
| 78 |
+
model: "KaniTTS Deutsch"
|
| 79 |
+
temperature: 0.6
|
| 80 |
+
top_p: 0.95
|
| 81 |
+
repetition_penalty: 1.1
|
| 82 |
+
max_len: 900
|
| 83 |
+
|
| 84 |
+
|
| 85 |
+
- text: >-
|
| 86 |
+
「いのち輝く未来社会のデザイン」というテーマが多くの人の心に残りました。
|
| 87 |
+
speaker_id: null
|
| 88 |
+
model: "KaniTTS Japanese"
|
| 89 |
+
temperature: 0.6
|
| 90 |
+
top_p: 0.95
|
| 91 |
+
repetition_penalty: 1.1
|
| 92 |
+
max_len: 900
|
| 93 |
+
|
| 94 |
+
- text: >-
|
| 95 |
+
¡Qué alegría volver a verte después de tanto tiempo!
|
| 96 |
+
speaker_id: "Ash"
|
| 97 |
+
model: "KaniTTS Español"
|
| 98 |
+
temperature: 0.6
|
| 99 |
+
top_p: 0.95
|
| 100 |
+
repetition_penalty: 1.1
|
| 101 |
+
max_len: 900
|
| 102 |
+
|
| 103 |
+
- text: >-
|
| 104 |
+
A veces, una simple mirada dice más que mil palabras.
|
| 105 |
+
speaker_id: "Nova"
|
| 106 |
+
model: "KaniTTS Español"
|
| 107 |
+
temperature: 0.6
|
| 108 |
top_p: 0.95
|
| 109 |
repetition_penalty: 1.1
|
| 110 |
+
max_len: 900
|
| 111 |
|
| 112 |
- text: >-
|
| 113 |
+
¿Será que todavía me recuerdas como antes?
|
| 114 |
+
speaker_id: "Ballad"
|
| 115 |
+
model: "KaniTTS Español"
|
| 116 |
+
temperature: 0.6
|
| 117 |
top_p: 0.95
|
| 118 |
repetition_penalty: 1.1
|
| 119 |
+
max_len: 900
|
| 120 |
|
| 121 |
- text: >-
|
| 122 |
+
조용한 밤에 혼자 있으니까 마음이 좀 이상해.
|
| 123 |
+
speaker_id: null
|
| 124 |
+
model: "KaniTTS Korean"
|
| 125 |
+
temperature: 0.6
|
| 126 |
top_p: 0.95
|
| 127 |
repetition_penalty: 1.1
|
| 128 |
+
max_len: 900
|
model_config.yaml
CHANGED
|
@@ -6,7 +6,54 @@ nemo_player:
|
|
| 6 |
|
| 7 |
models:
|
| 8 |
|
| 9 |
-
"KaniTTS":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
model_name: nineninesix/kani-tts-370m
|
| 11 |
device_map: auto
|
| 12 |
speaker_id:
|
|
@@ -26,11 +73,11 @@ models:
|
|
| 26 |
"Karim (AR)": karim
|
| 27 |
"Nur (AR)": nur
|
| 28 |
|
| 29 |
-
"Base Model v.0.
|
| 30 |
-
model_name: nineninesix/kani-tts-
|
| 31 |
device_map: auto
|
| 32 |
|
| 33 |
-
"Base Model v.0.
|
| 34 |
-
model_name: nineninesix/kani-tts-450m-0.
|
| 35 |
device_map: auto
|
| 36 |
|
|
|
|
| 6 |
|
| 7 |
models:
|
| 8 |
|
| 9 |
+
"KaniTTS English":
|
| 10 |
+
model_name: nineninesix/kani-tts-400m-en
|
| 11 |
+
device_map: auto
|
| 12 |
+
speaker_id:
|
| 13 |
+
"Andrew": andrew
|
| 14 |
+
"Katie": katie
|
| 15 |
+
|
| 16 |
+
"KaniTTS Chinese":
|
| 17 |
+
model_name: nineninesix/kani-tts-400m-zh
|
| 18 |
+
device_map: auto
|
| 19 |
+
speaker_id:
|
| 20 |
+
"Ming (Mandarin)": ming
|
| 21 |
+
"Mei (Cantonese)": mei
|
| 22 |
+
|
| 23 |
+
|
| 24 |
+
"KaniTTS Arabic":
|
| 25 |
+
model_name: nineninesix/kani-tts-400m-ar
|
| 26 |
+
device_map: auto
|
| 27 |
+
|
| 28 |
+
|
| 29 |
+
"KaniTTS Japanese":
|
| 30 |
+
model_name: nineninesix/kani-tts-370m-expo2025-osaka-ja
|
| 31 |
+
device_map: auto
|
| 32 |
+
|
| 33 |
+
|
| 34 |
+
"KaniTTS Deutsch":
|
| 35 |
+
model_name: nineninesix/kani-tts-400m-de
|
| 36 |
+
device_map: auto
|
| 37 |
+
speaker_id:
|
| 38 |
+
"Bert": bert
|
| 39 |
+
"Thorsten (Hessisch)": thorsten
|
| 40 |
+
|
| 41 |
+
|
| 42 |
+
"KaniTTS Español":
|
| 43 |
+
model_name: nineninesix/kani-tts-400m-es
|
| 44 |
+
device_map: auto
|
| 45 |
+
speaker_id:
|
| 46 |
+
"Ash": ash
|
| 47 |
+
"Nova": nova
|
| 48 |
+
"Ballad": ballad
|
| 49 |
+
|
| 50 |
+
|
| 51 |
+
"KaniTTS Korean":
|
| 52 |
+
model_name: nineninesix/kani-tts-400m-ko
|
| 53 |
+
device_map: auto
|
| 54 |
+
|
| 55 |
+
|
| 56 |
+
"KaniTTS multilingual":
|
| 57 |
model_name: nineninesix/kani-tts-370m
|
| 58 |
device_map: auto
|
| 59 |
speaker_id:
|
|
|
|
| 73 |
"Karim (AR)": karim
|
| 74 |
"Nur (AR)": nur
|
| 75 |
|
| 76 |
+
"Base Model v.0.3":
|
| 77 |
+
model_name: nineninesix/kani-tts-400m-0.3-pt
|
| 78 |
device_map: auto
|
| 79 |
|
| 80 |
+
"Base Model v.0.2":
|
| 81 |
+
model_name: nineninesix/kani-tts-450m-0.2-pt
|
| 82 |
device_map: auto
|
| 83 |
|