Offline Speech to Text converts recordings and audio files into text on your device using Whisper. Download a model from Models, choose the speech language, then record audio or import a file. Review the transcript and save or share it when you choose.

After a model is downloaded, transcription does not require uploading audio to a server. The initial model download requires internet access. Each recording or imported file is limited to five minutes.


Persian and English interfaces and light and dark themes are available. Requires Android 10 or later and a 64-bit ARM processor. Speed and accuracy depend on the model, device, language and audio quality. Review generated text before relying on it.