Whisper-based awesomeness!
Great solution
Please, add SRT file export option…
Thank you for the excellent WhisperBoard app — it works perfectly offline on my iPad, the transcription is very accurate, and the interface is clean and simple. I especially love that everything runs on-device, no ads, no subscriptions.
I often use it to record thoughts/interviews and I see segments with timestamps right in the app (e.g. [00:00:00.00 - 00:00:02.00]). This is already almost ready for subtitles!
Are you planning to add export to .srt (or .vtt) with these timestamps? It would be a huge help for creating subtitles for videos/audio. Many users have requested this in reviews, and it would make the app even more versatile.
Thank you for your great work! Looking forward to updates 😊
Best regards
Amazing !
Great App but SRT output Please
One Caveat
Actually a good app
The best app I could find
First time user
Best Transcription app, hands down
The Whisper models are free and open source, and you have access to 21 different Whisper models, all tuned for varying degrees of quality and performance. Even the smallest model works great. The bigger models run fine on my iPhone 14 but I’m limited by RAM. The biggest model should run perfectly on my M2 iPad Air.
Genuinely amazing work, thank you so much.
the best !
非常好用
Perfect app for private VoT
Exactly what I was looking for
Ничесебе!
Não é bom para outras línguas
I look forward for whisper ios keyboard!
Genial, Genau und dabei lokal ausgeführt!
Auf jeden Fall nochmals Danke!!!! Toll!!!
It’s the best and it’s free. It doesn’t get much better than that.
You can load audio files to transcribe or you can recordings directly in the app. I really couldn't ask for much more here.
A note for users who aren't familiar with the Whisper framework: you have to download models for the app. If you go into settings, you can select the model you want to use and thr app will download it. It comes with the ‘tiny’ model, which is fast but not as accurate as the larger models. If you want very high accuracy and are willing to have a slow transcribe rate, then I would recommend downloading a larger model though the settings.
I’m really thankful to the dev for making this. Absolutely fantastic.
(This review was written with speech to text using WhisperBoard)
