Speech-to-Text Transcription

it was an audio recording, transcribed with speech to text models. there's definitely some errors and words lost. I also tried to emphasize this

thanks for the feedback. we're trying to implement transcriptions.

(Can't edit anymore, but I meant "automatic transcription" above..)

I sometimes feel like its a rough transcript of the audio or similar. Because not only is the response apparently natively audio, but I've often seen the text adhere to the audio only as a best effort rather than a full on accurate representation of the audio.

"[...] by transcript, not waveform".

It's a direct transcript that I made using Descript, hadn't considered how odd it would be to read on its own! I'll fix it ha ha

Interesting idea. I'd be curious to hear if they turn out a good transcript or whether their lack of understanding of the subject material makes it garbled. I guess with enough people doing it, you can check out the most common transcriptions and go with those...

Rev works really well but an actual human that transcribes what you say and it takes a day.

It's the transcription, I think this kind of site should allow you to fall back to listening to the audio. Even with high accuracy, if there's no human curation, there will be pretty awkward mistakes every once in a while in the text.

For those of us that struggle with his cadence of speech, are transcripts available online?