Openai:whisper
Openai:whisper
Openai:whisper
Sign in Sign up
MIT license
data fix typo data/README.md (#2433) last week
Activity
notebooks Use ndimage.median_filter instead of signal.medfilter (#8… last year Custom properties
71.5k stars
tests large-v3 (#1761) last year
581 watching
whisper Add option to carry initial_prompt with the sliding window… 3 weeks ago 8.5k forks
.flake8 apply formatting with black (#1038) last year Report repository
.gitattributes fix github language stats getting dominated by jupyter no… last year
Releases 12
requirements.txt Relax triton requirements for compatibility with pytorch 2.… 2 months ago
setup.py Relax triton requirements for compatibility with pytorch 2.… 2 months ago
Whisper
[Blog] [Paper] [Model card] [Colab example]
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
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:
Alternatively, the following command will pull and install the latest commit from this repository, along with its
Python dependencies:
To update the package to the latest version of this repository, please run:
It also requires the command-line tool ffmpeg to be installed on your system, which is available from most
package managers:
# on Ubuntu or Debian
sudo apt update && sudo apt install ffmpeg
# on Arch Linux
sudo pacman -S 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:
There are six model sizes, four with English-only versions, offering speed and accuracy tradeoffs. Below are the
names of the available models and their approximate memory requirements and inference speed relative to the
large model. The relative speeds below are measured by transcribing English speech on a A100, and the real-world
speed may vary significantly depending on many factors including the language, the speaking speed, and the
available hardware.
Size Parameters English-only model Multilingual model Required VRAM Relative speed
The .en models for English-only applications tend to perform better, especially for the tiny.en and base.en
models. We observed that the difference becomes less significant for the small.en and medium.en models.
Additionally, the turbo model is an optimized version of large-v3 that offers faster transcription speed with a
minimal degradation in accuracy.
Whisper's performance varies widely depending on the language. The figure below shows a performance
breakdown of large-v3 and large-v2 models by language, using WERs (word error rates) or CER (character
error rates, shown in Italic) evaluated on the Common Voice 15 and Fleurs datasets. Additional WER/CER metrics
corresponding to the other models and datasets can be found in Appendix D.1, D.2, and D.4 of the paper, as well
as the BLEU (Bilingual Evaluation Understudy) scores for translation in Appendix D.3.
Command-line usage
The following command will transcribe speech in audio files, using the turbo model:
The default setting (which selects the turbo model) works well for transcribing English. To transcribe an audio
file containing non-English speech, you can specify the language using the --language option:
whisper --help
Python usage
import whisper
model = whisper.load_model("turbo")
result = model.transcribe("audio.mp3")
print(result["text"])
Internally, the transcribe() method reads the entire file and processes the audio with a sliding 30-second
window, performing autoregressive sequence-to-sequence predictions on each window.
import whisper
model = whisper.load_model("turbo")
# make log-Mel spectrogram and move to the same device as the model
mel = whisper.log_mel_spectrogram(audio).to(model.device)
More examples
Please use the Show and tell category in Discussions for sharing more example usages of Whisper and third-
party extensions such as web demos, integrations with other tools, ports for different platforms, etc.
License
Whisper's code and model weights are released under the MIT License. See LICENSE for further details.
© 2024 GitHub, Inc. Terms Privacy Security Status Docs Contact Manage cookies Do not share my personal information