---
dataset_info:
  features:
    - name: audio_path
      dtype: string
    - name: text
      dtype: string
  splits:
    - name: train
      num_bytes: 2726
      num_examples: 10
  download_size: 10015616
  dataset_size: 10015616
tags:
- voxtral
- asr
- speech-to-text
- fine-tuning
- audio-dataset
---

# Voxtral ASR Dataset

This dataset was created using the Voxtral ASR Fine-tuning Interface.

## Dataset Structure

- **audio_path**: Relative path to the audio file (stored in `audio/` directory)
- **text**: Transcription of the audio

## Dataset Statistics

- **Number of examples**: 10
- **Audio files uploaded**: 10
- **Total dataset size**: 10,015,616 bytes

## Usage

```python
from datasets import load_dataset, Audio

# Load dataset
dataset = load_dataset("Tonic/voxtral-dataset-20250913_174653")

# Load audio data
dataset = dataset.cast_column("audio_path", Audio())

# Access first example
print(dataset[0]["text"])
print(dataset[0]["audio_path"])
```

## Loading with Audio Decoding

```python
from datasets import load_dataset, Audio

# Load with automatic audio decoding
dataset = load_dataset("Tonic/voxtral-dataset-20250913_174653")
dataset = dataset.cast_column("audio_path", Audio(sampling_rate=16000))

# The audio column will contain the decoded audio arrays
audio_array = dataset[0]["audio_path"]["array"]
sampling_rate = dataset[0]["audio_path"]["sampling_rate"]
```