Practical overview
When it fits
Use Audiotext when you want a desktop workflow for transcribing recorded speech and then editing, translating, summarizing or exporting the transcript.
Who it is for
- People who need transcripts from recordings, videos, YouTube or linked media.
- Users preparing translated transcripts, summaries or subtitle files.
- Developers who want to build on or run the Python project from source.
Documented uses
- Turn media files, URLs, YouTube videos or microphone recordings into text.
- Process multiple files from a folder or monitor a folder for new files.
- Review, search, correct and play transcript segments.
- Translate or summarize transcripts and export subtitles or other text formats.
Requirements & limits
The repository identifies the project as open source under the MIT license. Release packages are listed for macOS 15 or later on Apple Silicon, 64-bit Windows and x86_64 Linux.
- Local WhisperX processing is distinct from using the Whisper or Google speech-to-text APIs.
- The README says WhisperX can run on CPU or use NVIDIA CUDA acceleration for faster processing.
- WhisperX and the Whisper API are described as supporting about 100 transcription languages with automatic language detection.
- The app offers multiple translation providers, including OpenAI, Claude, Gemini, DeepSeek, Mistral, Grok, Ollama, DeepL and Google Translate.
- The release installation includes FFmpeg, while building from source has separate system and Python requirements.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Install Audiotext from a release
Install the desktop app using the package for your operating system.
Before you start
- A supported system: macOS 15 or later on Apple Silicon, 64-bit Windows, or x86_64 Linux.
Steps
Download the package for your system from the latest GitHub release.
On macOS, open the DMG and drag Audiotext into Applications; allow it through Privacy & Security on first launch.
On Windows, run the installer.
On Linux, extract the archive and run ./install.sh.
Expected outcome
The Audiotext desktop app is installed.
Guide limitations
- The macOS app is not notarized; Windows may display a SmartScreen warning because the installer is unsigned.
- The installer offers CUDA acceleration on Windows and Linux when an NVIDIA GPU is present.
Prepare a local development environment
Set up the project and its dependencies from source.
Before you start
- FFmpeg.
- PortAudio on Linux for microphone recording and audio playback.
- uv and Python 3.10–3.13; uv can download a supported Python version.
Steps
Install the required system dependencies.
Clone the repository and change into the audiotext directory.
Install uv if it is not already available.
Run uv sync to create the virtual environment and install project dependencies.
Expected outcome
The project dependencies are installed in a local virtual environment.
Guide limitations
- The source setup installs development dependencies.
- On Linux and Windows, uv sync installs PyTorch with CUDA support, which the README notes is a several-gigabyte download.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 13 October 2026
- Original website title
- HenestrosaDev/audiotext · Official homepage title · retained in its source language