This page documents production updates to Cloud Speech API. You can periodically check this page for announcements about new or updated features, bug fixes, known issues, and deprecated functionality.
August 10, 2017
Time offsets (timestamps) are now available. Set the
enableWordTimeOffsets parameter to true in your request
configuration and the Cloud Speech API will include
time offset values for the beginning
and end of each spoken word that is recognized in the audio for your
request. For more information, see Time offsets (timestamps).
The Cloud Speech API has added recognition support for 30 new languages. For a complete list of all supported languages, see Language Support.
The limit on the length of audio that you can send with an asynchronous recognition request has been increased from ~80 to ~180 minutes. For information on Cloud Speech API limits, see Usage Limits. For information on asynchronous recognition requests, see
April 18, 2017
Release of the Cloud Speech API v1.
v1beta1 release of the Cloud Speech API has been
v1beta1 continues to be available for
a period of time as defined in the terms of service.
To avoid being impacted when the
is discontinued, replace references to
v1beta1 in your code
v1 and update your code with valid
API names and values.
language_code is now required with requests to the Cloud Speech API.
Requests with a missing or
language_code will return an error. (Pre-release versions of the
en-US if the
language_code was omitted from the request.)
SyncRecognize is renamed to
is renamed to
v1/speech:recognize. The behavior is unchanged.
AsyncRecognize is renamed to
is renamed to
The behavior is unchanged except that the
LongRunningRecognize method now
supports all of the
enum values. (Pre-release versions only supported the
sample_rate field has been renamed to
The behavior is unchanged.
EndpointerType enum has been renamed to
SpeechEventType enums have been removed.