README.md

---
title: PureVoiceAISpace
emoji: 💻
colorFrom: red
colorTo: blue
sdk: gradio
sdk_version: 6.26.0
python_version: 3.12.12
app_file: app.py
pinned: false
license: apache-2.0
short_description: AI speech logging transcription pipeline
---

# PureVoiceAI

An end-to-end audio processing pipeline designed to improve the quality of transcription by addressing the speech overlap problem. PureVoiceAI integrates machine learning models to perform audio source separation, voice activity detection, and speech transcription, turning noisy, overlapping audio into clean, attributed text.

## Key Features

- **Advanced Source Separation:** Utilizes custom trained model **RE-SepFormer** to isolate individual voices from noisy or overlapping audio tracks.
- **Voice Activity Detection:** Accurately determines "who spoke when," ensuring that transcribed text is correctly attributed to individual speakers.
- **Speech Transcription:** Converts the cleaned, separated audio streams into highly accurate text given voice activity timestamps.
- **Toolkits:** Built on top of audio processing toolkit [SpeechBrain](https://speechbrain.github.io/).

## Tech Stack

- **Language:** Python
- **Package Management:** pip
- **Audio Frameworks:** SpeechBrain
- **Core ML Models:** RE-SepFormer, pyAnnote, OpenAI Whisper

## References

This project heavily relies on the following toolkits and foundational models. If you use or build upon this repository, please consider citing the original authors:

**SpeechBrain (General-Purpose Speech Toolkit)**
@misc{speechbrainV1,
title={Open-Source Conversational AI with {SpeechBrain} 1.0},
author={Mirco Ravanelli and Titouan Parcollet and Adel Moumen and Sylvain de Langen and Cem Subakan and Peter Plantinga and Yingzhi Wang and Pooneh Mousavi and Luca Della Libera and Artem Ploujnikov and Francesco Paissan and Davide Borra and Salah Zaiem and Zeyu Zhao and Shucong Zhang and Georgios Karakasidis and Sung-Lin Yeh and Pierre Champion and Aku Rouhe and Rudolf Braun and Florian Mai and Juan Zuluaga-Gomez and Seyed Mahed Mousavi and Andreas Nautsch and Xuechen Liu and Sangeet Sagar and Jarod Duret and Salima Mdhaffar and Gaelle Laperriere and Mickael Rouvier and Renato De Mori and Yannick Esteve},
year={2024},
eprint={2407.00463},
archivePrefix={arXiv},
primaryClass={cs.LG},
url={https://arxiv.org/abs/2407.00463},
}
@misc{speechbrain,
title={{SpeechBrain}: A General-Purpose Speech Toolkit},
author={Mirco Ravanelli and Titouan Parcollet and Peter Plantinga and Aku Rouhe and Samuele Cornell and Loren Lugosch and Cem Subakan and Nauman Dawalatabad and Abdelwahab Heba and Jianyuan Zhong and Ju-Chieh Chou and Sung-Lin Yeh and Szu-Wei Fu and Chien-Feng Liao and Elena Rastorgueva and François Grondin and William Aris and Hwidong Na and Yan Gao and Renato De Mori and Yoshua Bengio},
year={2021},
eprint={2106.04624},
archivePrefix={arXiv},
primaryClass={eess.AS},
note={arXiv:2106.04624}
}

**RE-SepFormer Model**
@inproceedings{dellalibera2024resourceefficient,
title={Resource-Efficient Separation Transformer},
author={Luca Della Libera and Cem Subakan and Mirco Ravanelli and Samuele Cornell and Frédéric Lepoutre and François Grondin},
year={2024},
booktitle={ICASSP 2024},
}