Potentially, it’s about the audio quality more than anything. If there are not multiple voices talking over each other or music playing over it then it has a good chance

Yes! It supports many languages, you’ve made me realise a gap in the messaging in terms of communicating that value.

It can support up to 99 but the subtitle generation quality tends to decreases with more obscure languages as they wouldn’t have been in the training data

Yeah I can relate haha, that is one of the things I questioned, if a desktop app was the best format for this app. I do feel the majority of people are drifting more towards mobile tbh so may have to make a companion app lol

Technically its still free since there is no card requirement but we can agree to disagree here

yeah, this comment was such a nitpick, hn! You know!

Thanks, I appreciate it!

True, I would say though from my personal experience there is a serious study element inherent to learning languages to a high level too. So it depends on what the user wants, there are gamified apps like Duolingo but the efficacy of it has been questioned over the years

By study I mean adding to Anki for example as a flashcard

Ahh nice, hope it goes well!

Yeah that is a very good point, it’s something I thought about a lot when positioning the app. Essentially there are 2 mitigations to this, one is that it’s not really ideal for beginners and mainly targeting intermediate learners who can recognise or at least question errors to a certain extent but don’t want to create the entire subtitle file themselves. There is an inline editor for them to fix the errors.

The other thing is the speech recognition models are very sensitive to the audio. If the audio quality is good then the model would do a good job, I’m not sure how much experience you have with Whisper Large but it is very capable on normal speech. The issues arise when there are many competing sounds overriding each other

That’s fair enough! I am curious tho, are there any issues you face while using the app that could be handled better? Or any features that would make it easier to use?

ASR in LR is down often from my experience.

I tried Language Reactor a couple of years ago and bounced off of it because of its absolutely atrocious support for traditional character Chinese. Its page literally says there is none on Netflix despite most their Chinese content being from Taiwan.

Migaku is flawed, but better. This tool seems like a much less polished version of Migaku or Lexirise with just a local media player and no ability to use it on YouTube, Netflix, Bilibili, etc.