Practical overview
When it fits
Use Livewhisper when you want microphone speech transcribed sentence by sentence in a terminal, or when you want the included assistant for basic spoken queries and media commands.
Who it is for
- People who want microphone dictation displayed as text in a terminal.
- Developers or Linux users exploring a Whisper-based voice-command assistant.
Documented uses
- Turn microphone speech into near-live terminal text.
- Ask a voice assistant for weather, date and time, jokes, or Wikipedia searches.
- Issue basic media-player commands by voice.
Requirements & limits
- Core transcription requires Whisper, NumPy, SciPy and sounddevice.
- The assistant additionally requires requests, pyttsx3, wikipedia, bs4, espeak and python3-espeak.
- Media controls require some form of noise or echo cancellation to work properly.
- Basic math and trivia may rely on Google's instant-answer snippets and may not always work.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Prepare LiveWhisper for microphone transcription
Set up the documented prerequisites for sentence-by-sentence terminal dictation.
Before you start
- An existing installation of OpenAI Whisper
- Python dependencies: numpy, scipy, and sounddevice
- A microphone
Steps
Install Whisper, numpy, scipy, and sounddevice in the Python environment you intend to use.
Ensure the microphone is available to sounddevice.
Use the repository’s
livewhisper.pyas the transcription entry point; the README describes its microphone capture and transcription behavior.
Expected outcome
The documented prerequisites and entry point for LiveWhisper’s microphone transcription are identified; the README describes output as pseudo-live sentence-by-sentence dictation in the terminal.
Guide limitations
- Audio is buffered when volume and frequency thresholds are met and sent to Whisper after silence is detected.
Prepare the Whisper Assistant and use its documented voice requests
Set up the assistant’s stated dependencies and use its documented activation and request patterns.
Before you start
- The LiveWhisper dependencies: Whisper, numpy, scipy, and sounddevice
- Additional Python dependencies: requests, pyttsx3, wikipedia, and bs4
- espeak and python3-espeak
Steps
Install the LiveWhisper dependencies and the assistant’s additional Python dependencies.
Provide espeak and python3-espeak, which the README lists as additional requirements.
Activate the assistant by saying its default name, “computer”; “hey computer” and “okay computer” are also documented activation phrases.
Ask for a documented feature such as weather, date and time, jokes, or Wikipedia searches; media requests include play, pause, next, previous, stop, and “what’s playing?”
Close the assistant with Ctrl+C or by saying its name followed by “terminate.”
Expected outcome
The README documents voice activation, supported request examples, media commands, and two ways to close the assistant.
Guide limitations
- Media controls require some form of noise or echo cancellation to work right.
- Basic math and simple trivia depend on Google instant-answer snippets and sometimes do not work.
- The README does not provide installation commands or a verified execution test.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 14 October 2026
- Original website title
- Nikorasu/LiveWhisper · Official homepage title · retained in its source language