---
dataset_info:
  features:
  - name: first_line
    dtype: string
  - name: audio
    dtype: audio
  splits:
  - name: train
    num_bytes: 2285150618.152
    num_examples: 21676
  download_size: 2332238384
  dataset_size: 2285150618.152
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
license: cc-by-nc-4.0
task_categories:
- automatic-speech-recognition
size_categories:
- 10K<n<100K
---
# 📚 Dataset Summary

This dataset contains paired audio-text data generated using:
- Text prompts from the [BridgeData V2](https://rail-berkeley.github.io/bridgedata/) dataset.
- Synthetic speech generated using [Coqui-TTS](https://github.com/coqui-ai/TTS).
- Total 21676 Audio data.
- Approximately 100 style of speakers.
- 61528 seconds ≈ 17 hours.

---

# 📁 Dataset Structure

Each entry in the dataset contains:
- `first_line`: Textual instruction or caption from BridgeData V2.
- `audio`: Corresponding TTS-generated `.wav` file.

---

# 📜 Licenses and Usage

### 🔹 BridgeData V2
- License: [CC BY-NC 4.0](https://creativecommons.org/licenses/by-nc/4.0/)
- Attribution: The original text data is sourced from BridgeData V2, developed by the RAIL lab at UC Berkeley.
- Restrictions: **Non-commercial use only**. Any derivative dataset must also respect this condition.

### 🔹 Coqui-TTS
- License: [Mozilla Public License 2.0](https://www.mozilla.org/en-US/MPL/2.0/)
- Attribution: The TTS system used to generate the synthetic speech is based on the Coqui-TTS open-source toolkit.
---

# ✅ Terms of Use

By using this dataset, you agree to the terms of the licenses listed above.  
Please make sure any downstream usage (e.g., fine-tuning models, publication) complies with the original licenses.

---

# 🙏 Citation

If you use this dataset, please cite the original BridgeData V2 and Coqui-TTS projects accordingly.