Josh Holloway
Max Truemont
Search results
Sep 21, 2022 · Whisper is a Transformer-based model that can transcribe and translate speech in multiple languages from a large and diverse web dataset. It is robust to accents, noise and technical language, and outperforms supervised SOTA on CoVoST2 to English translation.
Whisper is an online community where you can share real thoughts, trade advice, and get the inside scoop. Chat with other users, join the 30 million+ people who use Whisper every month, and see the world in a new way.
- Social Networking
- 17+
- WhisperText LLC.
- Overview
- Approach
- Setup
- Available models and languages
- Command-line usage
- Python usage
- More examples
- License
- GeneratedCaptionsTabForHeroSec
[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.
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 stage...
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:
There are five 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; actual speed may vary depending on many factors including the available hardware.
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.
The following command will transcribe speech in audio files, using the medium model:
The default setting (which selects the small model) works well for transcribing English. To transcribe an audio file containing non-English speech, you can specify the language using the --language option:
Adding --task translate will translate the speech into English:
Run the following to view all available options:
Transcription can also be performed within Python:
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.
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.
Whisper's code and model weights are released under the MIT License. See LICENSE for further details.
Whisper is a Transformer-based model that can perform multilingual speech recognition, speech translation, and language identification. It is trained on a large dataset of diverse audio and can be installed and used with Python and ffmpeg.
2 days ago · Whisper is an online community where millions of people around the world share real thoughts, trade advice, and get the inside scoop. “The experience of spending time on Whisper is unlike any...
People also ask
What is whisper & how does it work?
What is Whisper app?
Is Whisper a robust speech recognition system?
Where can I buy a WhisperText app?
Whisper is a proprietary mobile app available without charge. It is a form of anonymous social media, allowing users to post and share photo and video messages anonymously, although this claim has been challenged with privacy concerns over Whisper's handling of user data.
Learn the meaning, synonyms, examples, and history of the word whisper, which can be a verb or a noun. Whisper can mean to speak softly, to make a sibilant sound, or to communicate a secret or a rumor.
Aug 28, 2023 · 1. Install Whisper from the App Store or Play Store. 2. Set up your notification settings. 3. Create your private profile. 4. Add friends from your contacts. 5. Browse the latest secrets and whispers nearby. 6. Tap Discover to search. 7. Tap a whisper to see its replies. 8. Tap Reply to reply.
- 76.7K
