The Speech-to-Text API now offers spoken punctuation and spoken emoji features. When you enable this feature, instances of spoken punctuation and emojis detected in your audio data will be replaced by the corresponding punctuation and emoji symbols. For example, if you send a request with spoken punctuation enabled, the transcript "how are you question mark" becomes "how are you?". See the Supported spoken punctuation page and Supported spoken emojis page for complete lists of spoken symbols available for your language.
The spoken punctuation and emoji features can be turned on by setting
Spoken punctuation is enabled by default for the
command_and_search model, but
can be disabled using the
enableSpokenPunctuation field. For all other
requests, spoken punctuation and spoken emoji are disabled by default.