Fact · Speech recognition
In openai-whisper, the language is left unset by default, and the program then detects the spoken language itself.
Checked against its source on October 17, 2026
In the source's own words
“language spoken in the audio, specify None to perform language detection”
whisper/transcribe.py: decoding defaults and command-line options (openai/whisper), OpenAI, GitHub (openai/whisper), 2026, Command-line options (--language).
Open the original- What it measures
- The default value of --language is None. Setting it, for example to Japanese, skips detection and forces that language.
- What it does not show
- Detection can pick the wrong language for short or noisy clips. Set the language yourself when you know it.
Source
Cited on
- How to Transcribe Audio to TextGuide
- Video to Text: Transcribe Videos and YouTubeGuide
- How to Run Whisper Locally on Your Own ComputerGuide
- Transcribe and Translate Audio in Other LanguagesGuide
- How do you translate a voice recording into English text?Question
- What are the best Whisper settings for accuracy?Question