Subtitle Edit 3.6.5 already supports speech recognition. English/Chinese/Russian/French/German's big model works well. It can be used to make timestamps and draft subtitles. Please download these model from this page( (https://alphacephei.com/vosk/models) and unzip it to the SE installation directory. first: select 'Video' -> 'Open video file' second: select 'Video' -> 'Video/audio to text...' -> 'Choose model' -> 'Generate' A two-hour movie takes about 30 minutes to generate subtitles. Please try it.

Thank you very much for information and instructions! Unfortunately, this feature is not (yet?) available for the Linux version :'( However, another member tried it with a few samples I gave him, and I think the results are very good. From my experience, much better than pyTranscriber. The only sample SubtitleEdit couldn't manage was the one with a song, but I think it's legit since no speech recognition engine I've tried could manage songs.* Once again, thank you 1170125006 !

Hi. I tested both English and German and the program did the subtitles. A little pointless, but if you work it, you can correct everything

I will have to give this a try when I have the time. Thanks!

Is there a way to use the voice recognition to translate Japanese to English, then use the auto translate feature to translate the resulting Japanese text into English?

Now any language recognition is not correct. If you don't know the original language, the translation will be bad, it will need to be corrected.