Supported features by language

This page describes the availability of Speech-to-Text features by language and locale.

Automatic punctuation

Cloud Speech-to-Text supports the enableAutomaticPunctuation field for en-US only.

Boost for speech adaptation

Using the boost field is only supported for en-US.

Class tokens for speech adaptation

Support for individual class tokens differs by language and locale. Review the list of supported class tokens for the full list.

Model selection

The standard phone_call, enhanced phone_call, and video models are only available for en-US.

Speaker diarization

Cloud Speech-to-Text supports speaker diarization for the en-US. Cloud Speech-to-Text only supports speaker diarization for transcribing phone calls—that is, when using the standard or enhanced phone_call models.