How to use it
- Pick your framework.
- Choose a model for speech recognition, synthesis or audio classification.
- Load it and fine-tune on your recordings.
How it's built
A curated markdown index of audio and speech pretrained models across frameworks, with descriptions and links.
Why it matters for businesses
Voice features (transcription, voice commands, audio tagging) ship faster when you start from a proven model.
In real life
A support team adds automatic call transcription by fine-tuning a pretrained speech model on its own call audio.



