Whisper openai/whisper

OpenAI's general-purpose speech recognition model for multilingual transcription, translation to English and language ID.

Creator kit + Compare GitHub ↗ Install guide (official) ↗

Whisper at a glance

Checked 15 Oct 2026

Whisper is OpenAI's open speech recognition model in six sizes (39M to 1550M parameters; README lists ~1 GB to ~10 GB VRAM). It transcribes many languages, translates speech to English and identifies the language, via a CLI or Python API. Code and weights are MIT. Install with pip; ffmpeg is required. It does not generate speech, clone voices or run agents.

Cite this pageRepoQuarry. “Whisper.” Checked 15 Oct 2026. https://repoquarry.dev/projects/whisper

How much of ElevenLabs does it replace?

1 of 4 jobs. Local transcription and translation to English; the README does not mention speaker diarization.

Speech generation Not covered

Voice cloning Not covered

Speech recognition Full replacement

Voice agents Not covered

For speech generation, voice cloning, voice agents, see other ElevenLabs alternatives. You may also need to build an interface, an API server and hosting yourself.

How much of Otter.ai does it replace?

1 of 3 jobs. Transcribes recordings locally; no live meeting capture, speaker labels or summaries.

Meeting transcription Partial replacement

Speaker identification Not covered

Meeting summaries Not covered

For speaker identification, meeting summaries, see other Otter.ai alternatives. You may also need to build an interface, an API server and hosting yourself.

Facts, with sources

Every answer links to where we found it. We read the documentation; we did not install it.

Report a correction

QuestionAnswerSourceChecked

Runs on my machine? Yes, CLI and Python package run on your own machine The following command will transcribe speech in audio files, using the turbo model github.com · whisper#command-line-usage ↗ 1d ago

Code licence MIT github.com · LICENSE ↗ 1d ago

Model weights licence MIT (weights released under the same licence as the code) Whisper's code and model weights are released under the MIT License. github.com · whisper#license ↗ 1d ago

Commercial use Yes, MIT code and weights Whisper's code and model weights are released under the MIT License. github.com · whisper#license ↗ 1d ago

Paid services needed? None required You can download and install (or update to) the latest release of Whisper with the following command github.com · whisper#setup ↗ 1d ago

Hardware Required VRAM per model from ~1 GB (tiny, base) to ~10 GB (large); turbo ~6 GB. CPU use is not discussed in the README Below are the names of the available models and their approximate memory requirements and inference speed relative to the large model. github.com · whisper#available-models-and-languages ↗ 1d ago

Operating systems README gives ffmpeg install commands for Ubuntu/Debian, Arch, macOS (Homebrew) and Windows (Chocolatey, Scoop) It also requires the command-line tool ffmpeg to be installed on your system, which is available from most package managers github.com · whisper#setup ↗ 1d ago

Setup effort Easy; pip install openai-whisper plus ffmpeg (Rust may be needed if tiktoken has no prebuilt wheel) pip install -U openai-whisper github.com · whisper#setup ↗ 1d ago