Fact · Speech recognition
The Whisper repository says its turbo model is an optimized version of large-v3 that transcribes faster, with minimal degradation in accuracy.
Checked against its source on October 17, 2026
In the source's own words
“the turbo model is an optimized version of large-v3 that offers faster transcription speed with a minimal degradation in accuracy”
openai/whisper: Robust Speech Recognition via Large-Scale Weak Supervision, OpenAI, GitHub (OpenAI repository), 2023, Model table, paragraph after it.
Open the original- What it measures
- The repository's model table lists turbo as a separate model, with about 8 times the relative speed of the large model.
- What it does not show
- Minimal degradation is OpenAI's own description. The repository does not give a turbo error rate in this passage, so compare turbo with large-v3 on your own audio.