Audiscript: Voice to Text

Audiscript turns speech into text entirely on your iPhone. No internet, no account, no subscription — your recordings never leave your device.

Most transcription apps upload your audio to a server. Audiscript does not. The AI model runs on your device's Neural Engine, so transcription works on a plane, in a basement, or with Airplane Mode switched on. Nobody — including us — can ever hear your recordings.

WHAT YOU CAN DO
• Record and transcribe in 32 languages, fully offline
• See who said what with automatic speaker labels (Pro)
• Search every word across all your recordings and jump straight to the moment it was said
• Tap any line to fix a name or a typo
• Organise recordings into colour-coded folders
• Export to TXT, SRT and VTT — free for everyone
• Export to Markdown and PDF (Pro)

BUILT ON WHISPER, RUNNING ON APPLE SILICON
Audiscript uses OpenAI's Whisper models converted to Core ML. A model ships inside the app, so transcription works from the very first launch — no download, no setup, no waiting. Want more accuracy? Download a larger model, up to next-generation Large V3 Turbo (Pro).

SUBTITLES IN ONE TAP
Every transcript carries timestamps, so SRT and WebVTT export is always one tap away — ready for YouTube, Premiere, Final Cut or any editor. Free, no watermark, no export limit.

PRIVACY, NOT AS A FEATURE — AS THE ARCHITECTURE
• No account, no sign-up, no email address
• Your audio is never uploaded, because there is no server to upload it to
• Everything works in Airplane Mode

FREE
• Unlimited recordings, up to 1 minute each
• Base and Small AI models
• TXT, SRT and VTT export
• Supported by ads

AUDISCRIPT PRO — ONE-TIME PURCHASE
• Record up to 2 hours
• Every AI model, including Large V3 Turbo
• Automatic speaker identification
• Markdown and PDF export
• No ads

No subscription. Pay once, keep it forever.

GOOD TO KNOW
Audiscript records audio inside the app. It does not import existing audio or video files. The app interface is in English.