Google Speech-to-text action - Convert short speech to text

This action allows you to convert a short audio file (less than 1 minute) into a text transcript.

Input fields

Field name	Description
Audio content	Accepts a content datapill, or audio content as a string. Workato will automatically convert the provided audio content into base-64 format.
Language	The language code of the audio file, as listed here (opens new window).
Encoding	Encoding of the audio file. Note that Google has not supported MP3 input yet. MP3 file must be converted to one of these supported encodings (opens new window) before inputting into this action.
Sample rate in Hertz	Sample rate of the audio file. Valid value range is 8000-48000. Usual optimal rate is 16,000, but varies across encodings and settings. Please refer to Google documentation (opens new window) to figure out the best sample rate.
Enable word time offsets	If true, the output includes a list of words, the start and end time offsets (timestamps) for those words.

Output fields

Field name	Description
Transcript	The extracted text transcript from the audio.
Confidence score	The confidence level for the entire transcription, with 0.0 as the lowest and 1.0 as the highest.
Words	List of words extracted from the audio. This list contains the output fields below.
- Start time	Time when the word appears in the audio.
- End time	Time when the word ends in the audio.
- Word	The extracted word.