DEDUPEDWho it is forJournalist or researcher with sensitive interview audio that cannot go to a cloud transcriber
The problemUploading private recordings to a hosted transcription service is a privacy risk.
The outcomeText transcripts of your audio, plus dictated text, produced entirely on your own hardware.
What you need first
- macOS, Windows or Linux desktop (Intel Mac runs the UI only and needs a remote backend)
- NVIDIA GPU (CUDA) or Apple Silicon (Metal) recommended; CPU-only works but is slower and needs about 5 GB free disk
- Git, Node.js 22+, Bun and Rust/Cargo only if building from source
- Speech models are downloaded on demand through the app; no paid keys or accounts required
Steps
- Install the app and open the transcription workspace.
- Load an audio or video file, or start the floating dictation widget.
- Install the speech model when prompted and run the transcription.
- Copy the transcript out and proofread it against the audio.
Prompt for your agent
Use debpalash/VoiceStudio to transcribe <audio file> locally. Return the transcript and verify a few spot-checked passages against the recording.
Caveat. The README lists transcription and dictation as features but does not publish accuracy figures, so always proofread names and numbers.
About VoiceStudio
A free local app for voice cloning, dubbing, transcription and audiobooks in 646 languages. Part of this week's top 10 repos: ranked #3 on Repo Radar. View debpalash/VoiceStudio on GitHub.
More with VoiceStudio
- Replace a paid cloud voice tool with a free local one Install the desktop app and generate cloned or designed voices on your own computer with no subscription or upload.
- Dub a video into another language with timed speech Feed a video to the dubbing workspace and get a translated voice track that lines up with the original timing.
- Give your AI agent a local text-to-speech tool through MCP Expose the local API and MCP server so a coding or chat agent can generate speech on demand.
- Run voice generation on an older laptop with no graphics card Use the CPU build to clone and generate voices on hardware without a dedicated GPU, trading speed for access.
- Turn a finished manuscript into an audiobook Use the audiobook workflow to render a long text into narrated audio with your chosen voice.
- Design a brand-new voice from a text description Describe the voice you want and generate it, with no reference recording needed.
- Batch-produce voiceovers from a watched folder Use reusable recipes and watch folders to render many scripts to audio without manual clicks.
Disclaimer
Deduped lists open-source repositories that belong to other people. We do not own, maintain, audit, or vouch for any of them, and a repo appearing here is not an endorsement or a safety check. The use cases, steps, and prompts are written by AI and may be incomplete, outdated, or just wrong. Where a use case is marked "Seen in the wild", we link to public posts where someone described something similar, but we have not verified those posts and the use case itself is still our AI's own write-up. Use all of this at your own risk. Only run code, commands, or agents on systems and accounts you are authorized to use, and read the code and every command before you run it, especially anything that touches credentials, files, or the network. Nothing here is professional, security, legal, or financial advice. All trademarks and project names belong to their respective owners, and their use here does not imply affiliation. Content is provided as is, without warranties of any kind. Found a mistake, a safety problem, or something you want removed? Email [email protected] and we will look at it promptly. Full disclaimer page