Fact · Transcription
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.
Checked against its source on October 17, 2026
In the source's own words
“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.”
RecognitionConfig: Speech-to-Text v1 REST reference, Google Cloud, 2026, encoding field description.
Open the original- What it measures
- Google's list of lossy codecs in the same passage includes MP3, OGG_OPUS and WEBM_OPUS, among others. The passage is a warning in the API reference for the encoding setting.
- What it does not show
- A general warning from the vendor, with no measured size of the effect. It applies to Google's recognizer and does not measure any other tool.