Official documentation
RecognitionConfig: Speech-to-Text v1 REST reference
Google Cloud, 2026
Read the originalRead on October 17, 2026
- What it is
- Google Cloud's reference for the configuration object sent with each Speech-to-Text v1 recognition request. It lists the audio encodings the API accepts, the channel counts for each, and a warning that lossy codecs and background noise can lower accuracy.
- How much weight it carries
- Official Google documentation, updated by Google as the API changes. It describes the API rather than measured accuracy, so it shows what the service accepts and warns about, not how much accuracy is lost.
Facts taken from this source
- Google says the accuracy of its speech recognition can be reduced when lossy codecs are used to capture or transmit audio, particularly if background noise is present.
“The accuracy of the speech recognition can be reduced if lossy codecs are used to capture or transmit audio, particularly if background noise is present.” (encoding field description)
- For multichannel recognition, Google's Speech-to-Text API accepts from 1 to 8 channels for OGG_OPUS audio, so OGG_OPUS is one of its supported audio encodings.
“Valid values for LINEAR16, OGG_OPUS and FLAC are 1 - 8” (audioChannelCount field description)