Hello @rawwerks, I appreciate your feedback! Regrettably, iOS doesn't allow long background tasks when the app is minimized. I'm considering trying out audio recording or playback during transcription to see if it continues to work when minimized. Theoretically, it should function because the app remains active while audio is being recorded or played, but this needs testing and there is high chance it would not pass App Store review process. Additionally, I'm already in the process of adding a custom cloud-based transcription feature using whisperX.

Regrettably, iOS doesn't allow long background tasks when the app is minimized. I'm considering trying out audio recording or playback during transcription to see if it continues to work when minimized.

Maybe something like having the app "read back" the transcription (text to speech) - which could be muted of course - just to have a "legitimate excuse" for letting the transcription run in the background. I don't know much about the App Store review process, but it seems to me that this audio playback would be a legitimate feature that they shouldn't have a problem with. (They don't need to know the true reason for the audio playback ;) )

Additionally, I'm already in the process of adding a custom cloud-based transcription feature using whisperX.

Awesome! I would definitely pay for this feature, assuming there was reasonable privacy/encryption/TOS in place.

@Saik0s I wonder why you'd offer an online version? Speed? Reliability? Otter is able to perform transcription while the app is in the background, e.g. https://www.andyibanez.com/posts/modern-background-tasks-ios13/

Is it what you tried and it doesn't work?

@ldenoue, I am integrating online transcriptions using large model, aiming for speed and quality. Additional cloud-based features, including text refinement via LLM, are also planned.

The background transcription you referenced is implemented already, but its reliability varies. The next update will introduce resumable transcriptions to improve situation with this issue.

Regrettably, iOS doesn't allow long background tasks when the app is minimized. I'm considering trying out audio recording or playback during transcription to see if it continues to work when minimized.

it works well,really amazing~

@Saik0s I appreciate how much work you've put into this, and how it is open source.

I know the cloud subscription would be a way to make money off of the app, and I think that makes sense.

I have my own dedicated Linux server for running machine learning models, and I would love to have a way to offload the transcription task from this app to that server. I don't think I would pay $4/mo to use my own server, but I would probably be willing to pay a $5 or $10 one-time fee to unlock that functionality.

So, while you're developing the cloud stuff... if you see an easy way to let the user configure the app to point to a custom server, and if you provided some general guidance of what kind of API to expose for the app to work correctly, I think that would be awesome, even if there was a small fee for that feature to help support this app.

It's your app, of course, so do whatever you think is best. I'm just mentioning something I was thinking about recently.