Google Speech To Text

Prev Next

Overview


The Google Speech To Text (GSTT) is a connected service that converts the audio speech to the written text.

Configuration


The configuration is similar to the other connecting services and there are the following steps to configure GSTT:

  1. Create or modify the module to have the following fields:

    008_Fields.PNG

  2. The webapp.yaml should contain the following records:

    001_Webapp.PNG

  3. If the webapp.yaml contains the proper records of GSTT - the corresponding connected service should be displayed in the list in the Site Configuration -> Connections -> Connected services:

    002_ConnectedServicesList.PNG

  4. Click the [+Add Service] button next to the GSTT item on the list.

  5. Create the Connected Account:

    003_CreateAccount.PNG

  6. Select the module mentioned in step #1:

    004_SelectModule.PNG

  7. Select the Source and Target fields:

    • Source -> MediaFile mediaholder;

    • Target -> TextBox field.

      005_SelecFields.PNG

  8. When saving the settings - the GSTT connected service should appear in the Connected Services list:

    006_CreatedConnectedService.PNG

  9. When the GSTT connected service is created - the following Action should be created in the module:

    007_Action.PNG

Google Speech To Text in use


When the configuration is done, the user can upload the audio file to the corresponding source field.
When the action is triggered - the target field will be automatically filled in by the written text taken from the audio file.

Notes and limitations in the current implementation


  • Currently video source files are not supported;

  • Currently only audio source files of the .WAV format are supported;

  • If the target text was manually edited and then the action is triggered - manual changes will not be saved.