openai /whisper
Robust Speech Recognition via Large-Scale Weak Supervision
★ 112,767⑂ 13,673👁 833⑃ 455● 0▤ 13
+69
日增长
+491
周增长
+2,383
月增长
112,767
stars
13,673
forks
833
watchers
455
prs
0
issues
13
releases
README
Whisper
[[Blog]](https://openai.com/blog/whisper)
[[Paper]](https://arxiv.org/abs/2212.04356)
[[Model card]](https://github.com/openai/whisper/blob/main/model-card.md)
[[Colab example]](https://colab.research.google.com/github/openai/whisper/blob/master/notebooks/LibriSpeech.ipynb)
Whisper is a general-purpose speech recognition model. It is trained on a large dataset of diverse audio and is also a multitasking model that can perform multilingual speech recognition, speech translation, and language identification.
Approach
!Approach
A Transformer sequence-to-sequence model is trained on various speech processing tasks, including multilingual speech recognition, speech translation, spoken language identification, and voice activity detection. These tasks are jointly represented as a sequence of tokens to be predicted by the decoder, allowing a single model to replace many stages of a traditional speech-processing pipeline. The multitask training format uses a set of special tokens that serve as task specifiers or classification targets.
Setup
We used Python 3.9.9 and PyTorch 1.10.1 to train and test our models, but the codebase is expected to be compatible with Python 3.8-3.11 and recent PyTorch versions. The codebase also depends on a few Python packages, most notably OpenAI's tiktoken for their fast tokenizer implementation. You can download and install (or update to) the latest release of Whisper with the following command:
pip install -U openai-whisper
Alternatively, the following command will pull and install the latest commit from this repository, along with its Python dependencies:
pip install git+https://github.com/openai/whisper.git
To update the package to the latest version of this repository, please run:
pip install --upgrade --no-deps --force-reinstall git+https://github.com/openai/whisper.git
It also requires the command-line tool ffmpeg to be installed on your system, which is available from most package managers:
bash
on Ubuntu or Debian
sudo apt update && sudo apt install ffmpeg
on Arch Linux
sudo pacman -S ffmpeg
on MacOS using Homebrew (https://brew.sh/)
brew install ffmpeg
on Windows using Chocolatey (https://chocolatey.org/)
choco install ffmpeg
on Windows using Scoop (https://scoop.sh/)
scoop install ffmpeg
You may need rust installed as well, in case tiktoken does not provide a pre-built wheel for your platform. If you see installation errors during the pip install command above, please follow the Getting started page to install Rust development environment. Additionally, you may need to configure the PATH environment variable, e.g. export PATH="$HOME/.cargo/bin:$PATH". If the installation fails with No module named 'setuptools_rust', you need to install setuptools_rust, e.g. by running:
bash
pip install setuptools-rust
Available models and languages
There are six model sizes, four with English-only versions, offering speed and accuracy tradeoffs.
Below are the names of the avaDiscussions
#2904ClapClip for Windows: offline Whisper transcription (whisper.cpp small / medium / large-v3-turbo) with batch folders0 comments · 2026-11-11#2901Murmury: Whisper listening to the other side of a call, answers from your own documents (macOS, open source)0 comments · 2026-11-05#2893BECKY0 comments · 2026-10-22#2888VoxFlow Pro: free push-to-talk Whisper dictation for macOS (Neural Engine to GPU fallback, stall watchdog)0 comments · 2026-10-16#28847 free in-browser Whisper base tools tested on an hour of audio: repeats and gaps at the 30-second seams0 comments · 2026-10-14#2883Speaker Caption Merger: combine per-speaker SRT/VTT locally while preserving overlaps0 comments · 2026-10-12#2876سلام0 comments · 2026-10-06#2875word_timestamps: punctuation inside a word starts a new "word" (`3.5%` -> ` 3`, `.5`, `%`; `"안녕"이라고` -> ` "안녕`, `"이라고`)1 comments · 2026-10-06
Latest release
v20250625 v20250625 2025-06-26相关报道 (6)
noteWhisperの論文は「文字起こしが完全」と言っている?公開コードと限界を読む2026-10-29Amazon Web Services (AWS)使用 SageMaker 对 Whisper 模型进行微调及部署2023-03-07智源社区OpenAI语音识别模型Whisper 2.0已经发布到GitHub2022-12-07OSCHINAOpenAI 开源语音识别模型 Whisper2022-09-26量子位能听懂口音的开源语音系统来了:OpenAI出品,支持99种语言,英文识别能力直逼人类2022-09-24InfoQ.cnOpenAI 宣布开源多语言语音识别系统 Whisper,英文识别能力接近人类水平2022-09-23
News & coverage (15)
Hacker NewsComplete silence is always hallucinated as "ترجمة نانسي قنقر" in Arabic▲ 568 · 340 comments · 2025-07-22Hacker NewsOpenAI quietly launched Whisper V2 in a GitHub commit▲ 335 · 126 comments · 2022-12-06Hacker NewsAsk HN: Real-time speech-to-speech translation▲ 158 · 70 comments · 2024-10-25Hacker NewsOpenAI releases Whisper v3, new generation open source ASR model▲ 117 · 58 comments · 2023-11-06Hacker NewsWhisper v3▲ 46 · 1 comments · 2023-11-07Hacker NewsShow HN: WhisperBuddy, Privacy-first AI-transcription app built after my layoff▲ 1 · 1 comments · 2025-06-08noteGetting Started with Free Transcription | How to Download and Use Whisper Desktop from GitHub2026-10-23MSNYou can’t ban a whisper: GitHub removes BitChat after government order, Section 79 powers come under scrutiny2026-07-26GIGAZINESummary of how to transcribe text from recorded files using voice recognition using OpenAI's 'Whisper' for free2022-09-28TechCrunchOpenAI open-sources Whisper, a multilingual speech recognition system2022-09-21