Fact · Dictation software
The Whisper repository describes Whisper as a general-purpose speech recognition model trained on a large dataset of diverse audio.
Checked against its source on October 16, 2026
In the source's own words
“Whisper is a general-purpose speech recognition model. It is trained on a large dataset of diverse audio”
openai/whisper: Robust Speech Recognition via Large-Scale Weak Supervision, OpenAI, GitHub (OpenAI repository), 2023, README, first paragraph.
Open the original- What it measures
- The project's public README, read on 2026-10-16.
- What it does not show
- The repository describes the model. The license is in the repository's license file and should be checked there before reuse.