TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Port of OpenAI's Whisper model in C/C++

8 pointsby pbowyerover 2 years ago

1 comment

pbowyerover 2 years ago
I&#x27;m sharing this because on my M1 Mac I couldn&#x27;t get Whisper to work with MPS. CPU transcribing was taking around 2x the audio time.<p>With Whisper.cpp with 4 threads and using the base.en model, a 40 minute podcast transcribed in 130 seconds.