Torchaudio is a library for audio and signal processing with PyTorch. It provides I/O, signal and data processing functions, datasets, model implementations and application components.
This code first pre-processes the audio files and trains an audio segment classifier, given a set of WAV files stored in folders (each folder represents a different class). Then, using the Telegram bot, she takes the voice of the person and says the name of the class
Install dependencies: pip install -r ./requirements.txt
Dataset contains voices from 11 classes
| Accuracy | |
|---|---|
| Test data | 0.9495 |