What problem it solves

Speech-processing workflows often require separate systems for transcription, translation, spoken language identification, and voice activity detection. A single multitask model can handle these tasks while reducing the number of stages in a traditional speech-processing pipeline.

Key features

• Multitask speech processing — One Transformer sequence-to-sequence model covers multilingual speech recognition, speech translation, spoken language identification, and voice activity detection.

  Source: README

• Multiple model sizes — The project provides six model sizes, including English-only and multilingual variants, with different speed and memory tradeoffs.

  Source: README

• Command-line transcription and translation — The command-line interface can transcribe audio files, specify a spoken language, and translate multilingual speech into English.

  Source: README

• Python and lower-level APIs — Python users can transcribe files directly or access lower-level functions for audio preparation, language detection, spectrogram generation, and decoding.

  Source: README

• Open-source release — The code and model weights are distributed under the MIT License.

  Source: README, LICENSE

Quick start

Copied verbatim from the project's README.

import whisper

model = whisper.load_model("turbo")
result = model.transcribe("audio.mp3")
print(result["text"])

Who it may suit

This project may suit developers who need speech transcription or translation from a command-line workflow, as well as Python users who need direct model access or lower-level audio processing.

Topics to discuss

• Which deployment scenarios would benefit most from the available model sizes?

• How should users choose between English-only and multilingual models?

• What community examples or integrations would make speech-processing workflows easier?

Written by ReadmeX AI from the project's README, source and public GitHub data; it may contain mistakes.

Read the original README https://github.com/openai/whisper#readme

ReadmeX opened this community for a trending project; it is not run by the project. Maintainers can claim it from the community page.