Transcriber AI – Free, end-to-end machine based transcription with speaker id

https://hello.transcriberai.com/

by WinH • 2 years ago

18 13 2 years ago

hello.transcriberai.com

jjbinx007• 2 years ago

Is there anything like this available to run locally? Our HR dept wants to use something to transcribe interviews but doesn't want to submit data to some random website.

broland• 2 years ago

I use whisper and pyannote (https://github.com/m-bain/whisperX), but it is a pain to run locally - I run it on a 4080. This seems to be actually trying to identify the speakers. Not sure what they are doing for that.

LunaRoot3• 2 years ago

seems like running it somewhere secured is the answer, rather than hosting on your own.

okeysmokey3• 2 years ago

Is this using Whisper or something else? It looks like the site performs multiple steps on the audio and did a decent job of guessing who was speaking.

WinH2• 2 years ago

Yes. We run run Whisper Large V3 (not Turbo) for the speech to text. It still seems to be the best open source model out there for that step. The main challenge we are trying to solve is Speaker Identification, which is a very time consuming process.

okeysmokey3• 2 years ago

How are you doing speaker id?

kerryritter• 2 years ago

Impressed by how accurately this identified and transcribed. Nice work

WinH2• 2 years ago

Thanks, did you try the editor?

sachinjarral1612• 2 years ago

Nice work. Like to learn more about this new technology

sachinjarral1612• 2 years ago

Nice work. AI will enhance this technology

LunaRoot3• 2 years ago

speaker identification will be the next breakthrough in AI transcription. Correctly identifying speakers greatly improves the end user's perceived quality of the final transcription product.

Transcriber AI – Free, end-to-end machine based transcription with speaker id

Related Stories