Speech Note is a Linux desktop and Sailfish OS application for taking, reading, and translating notes with integrated offline speech technology. It combines speech-to-text, text-to-speech, and machine translation in a single interface, allowing users to dictate notes, listen back to them, and translate them without ever sending data to the cloud. All processing is done locally, which means audio, text, and translations never leave the device, emphasizing strong privacy guarantees. The application supports multiple STT engines such as Coqui STT (DeepSpeech fork), Vosk, whisper.cpp, Faster Whisper, and april-asr, giving users flexibility in accuracy, speed, and hardware requirements. For text-to-speech, it can plug into a wide range of engines including espeak-ng, MBROLA, Piper, RHVoice, Coqui TTS, Mimic 3, WhisperSpeech, Kokoro, Parler-TTS, F5-TTS, and even classic S.A.M., making it highly customizable in terms of voices and languages.
Features
- Offline speech-to-text, text-to-speech, and machine translation in one app
- Local-only processing with no network requirement for privacy-sensitive use
- Support for multiple STT backends like Coqui STT, Vosk, Whisper, and more
- Integration with many TTS engines for diverse voices and languages
- Built-in model browser to download and manage language and speech models
- Available as Flatpak with Linux desktop and Sailfish OS support