Sussurro: Speech to Text AI

Transcribe audio to text with complete privacy, powered by the most advanced AI.

Sussurro (Italian for Whisper) lets you convert audio to text using cutting-edge Artificial Intelligence models, entirely on your device. No data leaves your Mac, iPhone, or iPad — no audio uploads, no transcriptions stored externally, and absolutely no analytics. Just seamless, secure transcription of great quality with total privacy.

WHY CHOOSE SUSSURRO?
- Complete Privacy: Everything happens on your device. Your audio and text stay yours, always.
- Powered by Whisper: Sussurro downloads and runs locally OpenAI’s state-of-the-art Whisper models for accurate speech-to-text transcription.
- Works On All Your Apple devices: the app works natively on Mac, iPhone, and iPad, and is optimised for Apple silicon and the Neural Engine for fast, efficient processing.

FLEXIBLE, FAST & ACCURATE
- Choose Your Model: Select from multiple AI models to balance speed and accuracy according to your needs. Sussurro makes model selection simple and easy to understand — no AI expertise required.
- Transcribe offline: Download models in-app and, once you get them the first time, you won't need any Internet connectivity.
- Supports audio and video files!

WORK YOUR WAY
- Load and transcribe existing Audio files.
- Capture audio directly within the app and transcribe it right away.
- When you're done transcribing, Copy, Save to file or Share text to any application.

TRY FOR FREE, BUY ONE TIME IF YOU LIKE IT.
- Free to Test: Preview any model with complete real-time transcription during processing. The free version shows the first 300 characters of the final transcription.
- One-Time Purchase: Unlock unlimited transcription capabilities across all your Apple devices with a single, affordable in-app purchase — NO SUBSCRIPTIONS, no hidden fees. Plus, Family Sharing support means your family members get full access, too.

WHAT ELSE YOU NEED TO KNOW:
- Sussurro Privacy Policy and Terms of use are written by a human (me, Cesare, hi!) to be read by other humans. And you should read it: https://cdf1982.com/privacy/sussurro_terms_of_service_and_privacy_policy
- Sussurro relies on OpenAI's Whisper models, which are distributed under the MIT license. However, OpenAI is neither involved nor affiliated with this app.
- Please be advised that Whisper models can be quite large, so you should download them only when you don't incur in costs from your provide.
- Transcription speed is affected by your device performance and the model you choose.
- Accuracy depends greatly on the source audio quality and the model chosen for transcription.
- Transcriptions may include errors, including significant mistakes.
- Sussurro does not separate different speakers and does not format the output text.

Download Sussurro today and transcribe with confidence — privately, efficiently, and forever.