Turn recordings into text on your iPhone. Download the models once, then transcribe offline.
Transcribes 99 languages · Audio is never uploaded · No account, no sign-in
The transcript is produced on the phone after you stop – watch the progress, then check it against the audio.
It detects the language by itself, or you set one in Settings – recognised on your device either way.
Copy it, or export Plain Text, Markdown, SRT, WebVTT or JSON – on every plan, for the text you have.
01
Tap record. Pause if you need to, or drop a flag on a moment worth finding later.
02
Recognition starts on the phone. The transcript appears when it is done.
03
Play the audio, follow the timecodes, and edit a segment where needed.
04
Plain Text, Markdown, SRT, WebVTT or JSON – on every plan.
A timecode on every segment. Play the audio, follow along, and open a segment to edit it where a word is wrong.
Audio already saved stays available. If a recording is interrupted, reopen Kaseta and follow the recording status.
No account and no sign-in. Your audio is never uploaded – the models are downloaded once, then it all runs on the phone.
HelloOláMerhabaПриветHalloCześćHolaBonjourCiaoHalloHejПривітこんにちは안녕하세요你好नमस्तेمرحباשלוםสวัสดีXin chàoHaloΓειά σου
No account and no upload – your recordings and their transcripts stay on the phone.
