How does the API handle files that contain spoken audio in multiple languages?

Our models are capable of transcribing audio files that contain spoken audio in more than one language. This is referred to as code switching. Code switching performance varies by language pair. Check out this section of our documentation to learn more.