/speech-to-text-utils

Speech to text CLI that helps you manage speech customizations.

Primary LanguageJavaScriptApache License 2.0Apache-2.0

Speech to Text CLI

Speech to text CLI that helps you manage speech customizations.

Getting Started

Make sure you read the documentation for Speech to Text before using this library.

npm install watson-speech-to-text-utils -g

Usage

$ watson-speech-to-text-utils

Usage:  <command> [options]


Commands:

  set-credentials [options]                  Set Speech to Text username and password
  base-models-list [options]                 List all the base models
  customization-create-and-train [options]   Create a customization model using a Conversation workspace JSON file
  customization-add-corpus [options]         Add a new corpus (i.e. workspace.json) to an existing customization
  customization-list [options]               List all the customization
  customization-train [options]              Train a customization
  customization-status [options]             Get customization status
  customization-delete [options]             Delete a customization
  customization-list-words [options]         List all out-of-vocabulary(OOV) words for a customization
  corpus-list [options]                      Lists information about all corpora for a customization model
  corpus-status [options]                    Lists information about all corpora for a customization model
  corpus-add-words [options]                 Add a group of words from a file into a corpus
  corpus-add-word [options]                  Add a single word to the corpus
  corpus-delete-word [options]               Delete a single word from a corpus
  corpus-from-workspace [options]            Build a Speech to text corpus from a Conversation workspace

Options:

  -h, --help     output usage information
  -V, --version  output the version number

Getting Started

All the commands need an Speech to Text username and password. Take a look at the Service credentials for Watson services page to learn how to get credentials for the Watson services.

How to get the workspace.json file

  • Navigate to your Bluemix console and open the Conversation service instance where you imported the workspace.
  • Click the menu icon in the upper-right corner of the workspace tile, and then select Download as JSON.

Adding multiple words to a customization

Usage: watson-speech-to-text-utils corpus-add-words [options]

  Add a group of words from a file into a corpus

  Options:

    -h, --help                                 output usage information
    -u, --username [username]                  Speech to text username
    -p, --password [password]                  Speech to text password
    -i, --customization_id <customization_id>  The customization identifier
    -w, --words <words>                        The JSON file with the words to add to the corpus

The words JSON file should look like:

{
  "words": [{
    "display_as": "could",
    "sounds_like": [ "could" ],
    "word": "culd"
  }, {
    "display_as": "closeby",
    "sounds_like": [ "closeby" ],
    "word": "closeby"
  }, {
    "display_as": "cya",
    "sounds_like": [ "cya", "see ya" ],
    "word": "cya"
  }]
}

Adding a single word to a customization

Usage: watson-speech-to-text-utils corpus-add-word [options]

  Add a single word to the corpus

  Options:

    -h, --help                                 output usage information
    -i, --customization_id <customization_id>  The customization identifier
    -w, --word <word>                          The word
    -s, --sounds_like [sounds_like]            The pronunciation of the word
    -d, --displays_as [displays_as]            How the word is displayed

For example:

watson-speech-to-text-utils corpus-add-word -i 67f35b00-8c0d-12e6-8ac8-6333954f158e -w cya -d cya -s "cya, see ya"

License

This sample code is licensed under Apache 2.0.

Contributing

See CONTRIBUTING.

Open Source @ IBM

Find more open source projects on the IBM Github Page