Overview
TrueScribe is a desktop transcription workspace for turning interviews, meetings, lectures, podcasts, videos, and voice memos into editable text. It runs on Windows and Linux and processes recordings locally by default, keeping media and transcripts on the user's machine.
The app combines transcription with timestamped editing, playback, speaker labeling, translation, transcript research, search, and export tools.
Key Features
- Transcribes imported audio and video files or dictation recorded inside the app.
- Aligns editable transcript segments with timestamps and waveform playback.
- Supports speaker detection, naming, and assignment to transcript segments.
- Provides local translation while preserving the original transcript.
- Runs reusable transcript research prompts for summaries or key points.
- Exports text, subtitle, document, and structured-data formats according to the selected plan.
How It Works
Add an audio or video file, or record a voice memo directly. Pro users can also use direct media links and supported public media pages, as well as queue multiple files for batch transcription.
TrueScribe runs Whisper locally and selects a model suited to the available hardware. The model is downloaded on first use and reused offline; both CPU and Vulkan runtimes are supported. After transcription, users can replay the recording, correct text with undo and redo, identify speakers, translate content, search the transcript, and export the result.
Privacy and Connectivity
Local files are decoded and transcribed without an upload or processing account. Network access is used for model downloads, media links, licence checks, updates, and error reporting.
If a cloud model or remote endpoint is selected for transcription, translation, or AI Research, the content needed for that operation is sent to the chosen provider. Offline use is available after the required models have been downloaded.
Pricing
The Free plan includes unlimited local-file transcription, available transcription models, dictation, history, playback, editing, search, AI Transcript Research, and TXT, SRT, and WebVTT export.
Pro costs US$64 as a one-time purchase for three devices. It adds batch transcription, media-link support, local AI translation to 25 languages, automatic speaker detection and assignment, custom research prompts, and extended export formats; applicable taxes are handled at checkout.