Support
Stuck on something? Most issues are solved below — and if yours isn't, email vilvaathiban@gmail.com and I'll get back to you personally, usually within two days.
Troubleshooting
The transcript is empty or wrong
The model works best on clear speech with little background noise, and it auto-detects the language. Very short notes (under two seconds), music, or heavily overlapping voices can produce poor results. If a clear recording transcribes badly, email the file — that usually points at a decoding issue that can be fixed.
It is slow on my phone
Transcription runs on the device's own processor; a 30-second note typically takes 5–15 seconds. Older iPhones take longer. The first transcription after launch also loads the model, which adds a few seconds once.
Which formats work?
.opus, .ogg, .m4a, .aac, .mp3, .wav, .caf, .aiff, .flac and most other audio files.
Which languages?
The bundled model recognises around 100 languages and detects the language automatically.
Feature requests
The App is actively developed and feature requests shape the roadmap. Email vilvaathiban@gmail.com with the subject "Feature Request".
Privacy
Everything the App does happens on your device — nothing is uploaded. Read the full privacy policy.