Last week, I needed a transcription of the latest episode of
. I for reals searched for some way to do it on my Mac. I came across
.
Whisper is an automatic speech recognition (ASR) system trained on 680,000 hours of multilingual and multitask supervised data collected from the web.
They actually open sourced it and when you look hard enough, you find out how to use it.
On Linux. Using terminal commands.
Some searches proved unsuccessful, so I used my work elevenlabs account, which I really hate to do. Work is work, personal is personal. Not a great solution. So, I wrote an app. It’s called Voxtext. You drag any audio or video onto it and click transcribe, and it generates text. On my M2 MacBook Air, it processed 70 minutes of audio in 30 minutes. It can do a 30 second video almost instantly.
And, it’s free. Yep. Free like beer. Sure, you can kick in a few buckets to help fund the development machine further, but if you need to transcribe any audio or video file, and you are tired of trying to find the way to do it, just download
. Mac, Windows, Linux.