[spudlyo]

So, this project consists of a ~175 line README and a ~500 line Python program that glues yt-dlp and Kroko together. Neat.

I guess if it encourages you to install and figure out how to use ffmpeg, yt-dlp, kroko, numpy, and onnx that's a good thing. Sometimes just knowing a thing is possible is a huge benefit.

[iririririr]

I see the value as a centralized anti-content-blocker.

This repo is now a good way to centralize hacks around the sure-to-come blockers those platforms will add to prevent download.

Just like uBlockOrigin was a way to centralize all the "just run this greasemonkey script" comments, I can see this getting a huge following for people who really value transcriptions.

[mrkn1]

I appreciate the perspective! higher ceiling than I'd put on it, but if it gets there awesome. PRs welcome!

[mrkn1]

thank you. You nailed the actual value, that's right. The real win is just knowing you can do this on a laptop CPU, offline, no GPU or cloud bill. There are tiny done-for-you details, like rescaling token timestamps back to real time after the atempo speedup so --timestamps doesn't lie to you, but they are minor.

[mscdex]

Why the choice of Kroko over something like parakeet-tdt-0.6b-v3, which is also faster than realtime on CPU?

[nshm]

Kroko models are more accurate and their size is just a hundred megabytes compared to parakeet (2.5 gigabytes in default fp32)

[mscdex]

Do you have a link to results confirming this? Kroko does not seem to be on the Open ASR Leaderboard. Parakeet has an average WER of 6.32 across several common datasets.

[mrkn1]

Kroko's website says benchmarks aren't formalized yet. FWIW, this url says 5% WER for English [0]. though it doesn't specify the dataset, so not directly comparable to Parakeet's 6.32 on the Open ASR Leaderboard

Best way to judge is to try it on your own audio

[0] https://huggingface.co/hudaiapa88/sherpa-stt-onnx