Skip to main content

Speech to text

A speech recognition software that allows you to recognize and translate spoken speech into text.

Speech to text

Installation​

Available by default in the DocSpace Saas solution.

You can enable it following the instructions here.

Configuration​

To start working with the plugin:

  1. Generate the API token using AssemblyAI.
  2. Go to Developer Tools -> Integration -> Plugins.
  3. Click Settings icon next to the Speech to text plugin.
  4. Enter the API token in the corresponding field.

Once the API token is enabled, the plugin becomes available to all users of the current DocSpace portal.

There you can also view the plugin metadata, such as the author, version, status, homepage, and description.

Usage​

To convert an audio or video file to text:

  1. Open the Documents section or one of the available rooms.
  2. Right-click a file name to open the file context menu.
  3. Click Convert to text.
  4. If speech is detected in the audio / video file, a new file called "Filename + text" will be created.

Plugin structure​

Repository on GitHub: speech-to-text.

All the required files are described here.

Interfaces​

The following plugin interfaces are used:

  • IPlugin. Required for each plugin. It contains the plugin status (PluginStatus) variable, used to embed the plugin into the DocSpace.
  • IApiPlugin. Required as we implement a third-party service (AssemblyAI).
  • ISettingsPlugin and ISettings. Used to add settings block for plugin's configuration. The users will access this block from Settings -> Integration -> Plugins to adjust the plugin's parameters.
  • IContextMenuPlugin and IContextMenuItem. Used to implement a context menu action.

Support​

If you want to request a feature or report a bug regarding this plugin, use the issues section on GitHub.