harvard-edge/multilingual_kws

Few-shot Keyword Spotting in Any Language and Multilingual Spoken Word Corpus

39
/ 100
Emerging

This project helps you quickly create a system to detect specific keywords spoken in any language, even if you only have a few examples of the keyword. You provide audio recordings and a small number of keyword examples, and it outputs a model that can identify those keywords in new speech. It's designed for researchers or product developers building voice-controlled interfaces or spoken language understanding systems.

186 stars. No commits in the last 6 months.

Use this if you need to spot specific keywords across many languages with minimal training data for each keyword.

Not ideal if you're looking for a general-purpose speech-to-text transcription service rather than specific keyword detection.

keyword-spotting voice-interfaces speech-recognition multilingual-audio spoken-language-understanding
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 8 / 25
Community 21 / 25

How are scores calculated?

Stars

186

Forks

42

Language

Jupyter Notebook

License

Last pushed

Dec 06, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/harvard-edge/multilingual_kws"

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