di37/speech-to-text-fine-tuning-on-unseen-language

This projects aims to show how whisper model can be fine-tuned on language it was not trained but is trained on similar language to it.

13
/ 100
Experimental

This project helps speech technologists and AI engineers adapt existing speech-to-text AI models to new, previously unsupported languages. You provide audio recordings and their corresponding text transcripts in the target language. The output is an improved speech-to-text model capable of transcribing the new language more accurately.

No commits in the last 6 months.

Use this if you need to create a high-performing speech-to-text model for a language that current general-purpose models don't support well, but is similar to a language they do support.

Not ideal if you need a speech-to-text solution for a widely supported language, or if you don't have existing audio and text data for your target language.

speech-recognition natural-language-processing AI-model-adaptation low-resource-languages machine-learning-engineering
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 8 / 25
Community 0 / 25

How are scores calculated?

Stars

11

Forks

Language

Jupyter Notebook

License

Last pushed

May 10, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/di37/speech-to-text-fine-tuning-on-unseen-language"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.