Fact · Transcription
The Whisper README says the turbo model is not trained for translation tasks, and that translating non-English speech into English needs one of the multilingual models: tiny, base, small, medium or large.
Checked against its source on October 17, 2026
In the source's own words
“the `turbo` model is not trained for translation tasks”
openai/whisper: README (raw text), OpenAI, GitHub (openai/whisper), 2026, Usage section, after the default command.
Open the original- What it measures
- Whisper's default model in the command-line tool is turbo, which the README describes as working well for transcribing English. The same paragraph names the multilingual models to use for translation.