Whisper: Transcribe or translate recorded speech
Added · Updated
Made by OpenAI
View profile on GitHub (opens in a new tab)Turn recorded speech into transcripts, translated English text, or a spoken-language label.
For developers and researchers who need local audio transcription.
What it does
Whisper is a general-purpose speech-recognition model that can transcribe speech, translate supported non-English speech into English, and identify spoken language. The repository provides a Python package and a command-line program for audio files. Its models offer different memory and speed trade-offs, and the source says performance can vary widely by language, speaking speed, and available hardware. The turbo model is optimized for transcription but is not trained for translation; the source directs translation users to multilingual model variants.
How to use it
Install openai-whisper and FFmpeg, then give the `whisper` command an audio file. Choose a multilingual model and the translate task for supported non-English speech that should become English text.
Common use cases
- Give the command an interview recording and receive a text transcript to review.
- Provide a Japanese audio file with the language and translate options to receive English text.
- Run the language-detection function on an audio segment and receive the most likely spoken-language label.
Can I use this?
Requirements Python and FFmpeg; the source documents pip installation and a command-line invocation. The repository is MIT licensed.
Caveat The project’s documentation cautions that accuracy varies across languages and that less-represented languages may have weaker results.
Worth knowing The source says performance varies widely by language; the default turbo model is not trained for translation, so multilingual model variants are needed for that task. FFmpeg is required. MIT.
Why it's here
Not trending this week
Last seen on the GitHub trending list on 21 October 2026. It stays on BuildTube so you can still find it.
- 112,230 people have starred it on GitHub.
Numbers checked on 8 November 2026; not refreshed since.
Behind it
View setup guideBuildTube has not run or verified this project. Everything above is written from what the creator published.
Made this?
Quick questions
- What is Whisper?
- Whisper is a general-purpose speech-recognition model that can transcribe speech, translate supported non-English speech into English, and identify spoken language.
- What can I use Whisper for?
- Give the command an interview recording and receive a text transcript to review; Provide a Japanese audio file with the language and translate options to receive English text; Run the language-detection function on an audio segment and receive the most likely spoken-language label
- Can I run Whisper on my computer?
- It runs from a terminal on your own computer.
- What do I need to use Whisper?
- Python 3.8–3.11 is the source’s expected range, a recent PyTorch setup, and the FFmpeg command-line tool.
- What license does Whisper use?
- The repository is MIT licensed.
More from this week
- agent-desktop Let AI agents read and click desktop apps
For developers building computer-use agents: agent-desktop lets an AI agent read and operate desktop apps on Mac, Windows or Linux from a command line.
- Auto-Ontology Give AI agents checked business meanings for your database
Help AI agents answer questions about company databases using reviewed definitions and SQL, with each answer showing its steps.
- Spec Kit Turn a written spec into plans, tasks and code
Spec Kit gives your AI coding agent a step-by-step process to build a feature, fix a bug or assess an idea.
Explore by topic
Post this
Download share imageFor makers
Made this? Show it’s on BuildTube — add the badge to your README or embed the card on your site.
README badge (Markdown)
[](https://buildtube.app/p/gh-openai-whisper/)Embed card (HTML)
<iframe src="https://buildtube.app/embed/gh-openai-whisper/" width="560" height="240" style="border:0;max-width:100%" loading="lazy" title="Whisper on BuildTube"></iframe>