Cloud Speech-to-Text transform

This document shows you how to use the Google Cloud Speech-to-Text transform plugin to convert audio files into text files.

Before you begin

Ensure that you have enabled the Speech-to-Text API.
Build a pipeline that reads data from Google Cloud Storage. Upload the following speech file (hello.wav) to a Cloud Storage bucket.

In the left navigation panel, navigate to Studio.
In the left panel, under the Source section, select a source. In this example, Google Cloud Storage.
Under the Transform section, select Google Cloud Speech-to-Text.
Under the Sink section, select a sink. In this example, Google Cloud Storage.
On the canvas, connect the three items.
Hover over the source, Google Cloud Storage. Click on the Properties button that appears.
In the Google Cloud Storage Properties window, set the Path to your Cloud Storage bucket path, and make sure the Format is “blob”.
Click the X button at the top right to save your changes.
Hover over the transform, Google Cloud Speech-to-Text. Click on the Properties button that appears.
Specify the Sampling Rate, Parts, and Text fields.
Click Get Schema, and then Apply, to automatically apply the output schema.
Click the X button at the top right to save your changes.
Hover over the sink, Google Cloud Storage. Click on the Properties button that appears. Set the Path to an output bucket.
Name your pipeline and click Deploy.
Click Run. It takes a few minutes for the pipeline to run.
Once the pipeline succeeds, you can view your transcribed text data in Cloud Storage or whichever sink you chose.