MingLunHan/CIF-ColDec

[ICASSP 2022] Improving End-to-End Contextual Speech Recognition with Fine-Grained Contextual Knowledge Selection

33
/ 100
Emerging

This project helps improve the accuracy of speech recognition systems, especially when dealing with specific names, technical terms, or domain-specific vocabulary. It takes audio input and a list of words or phrases that are likely to appear in the speech, then produces a more accurate transcription by prioritizing those contextual clues. This is useful for anyone working with automated speech-to-text conversion where precise recognition of certain terms is critical.

No commits in the last 6 months.

Use this if your automated speech recognition (ASR) system frequently misinterprets proper nouns, unique product names, or jargon relevant to your industry.

Not ideal if you are looking for a general-purpose ASR solution without the need for specific contextual biasing, or if you don't have defined lists of terms to guide the recognition.

speech-to-text audio-transcription meeting-minutes voice-assistant call-center-analytics
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 10 / 25

How are scores calculated?

Stars

25

Forks

3

Language

License

Apache-2.0

Last pushed

May 18, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/MingLunHan/CIF-ColDec"

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