search

krikristoophe / whisper_flutter_plus

Ready to use whisper.cpp models implementation for iOS and Android
https://pub.dev/packages/whisper_flutter_plus
MIT License
18 stars 11 forks source link

Detect Language + Options + Playback sync Example #6

Open jtkeyva opened 1 year ago

jtkeyva commented 1 year ago

Hello,

Does the package have all the functionality Whisper offers? Is there a way to manually set the options to detect language, generate karaoke and set segment length? What about extracting audio from video and transcribing?

options such as these would be super useful if not so much trouble to implement:

-owts, --output-words [false ] output script for generating karaoke video -ml N, --max-len N [0 ] maximum segment length in characters -l LANG, --language LANG [en ] spoken language ('auto' for auto-detect)

What would be a really cool example would be to playback the audio or video in sync with the transcribed words possibly in karaoke mode and/or with closed captions. Perhaps video transcription is beyond the scope of the example but it sounds possible to do.

With Flutter, the video_player package has support for the SubRip and WebVTT formats for captions. These files will be parsed to a ClosedCaptionFile that can be interpreted by the video_player package.

Happy to buy you a coffee for your efforts if you have a link.

Cheers

tempo-riz commented 5 months ago

hey @jtkeyva did you find a solution for this ? I'm looking to the language auto select feature :)