Glossary → Transcription & speech recognition
Transcription & speech recognition
Code-switching
Code-switching is when a speaker alternates between two or more languages within a conversation or a single sentence — 'Spanglish', Hindi-English, Tagalog-English — and it is one of the hardest real-world cases for speech recognition.
Multilingual models handle code-switching far better than monolingual ones because they were trained on many languages in one network, but rapid mid-sentence switches still deserve a review pass.
Real recordings code-switch constantly — business meetings in the Gulf, family recordings in diaspora communities, tech talks everywhere — so a model's code-switching behaviour matters more than its single-language benchmark.
Related terms
Multilingual speech recognitionMultilingual speech recognition uses one model trained across many languages — Whisper's published list covers…
Automatic speech recognition (ASR)Automatic speech recognition (ASR) is the technology that converts spoken language in audio into machine-reada…
Put the term to work
Transcribe audio or video in 99 languages — speaker labels, word timestamps, captions.
Transcribe a file →