Inworld speech provider

An Inworld speech provider lets Live Hub use your Inworld account for text-to-speech (TTS).

Adding the provider requires the API key from your Inworld account. Inworld's advanced configuration parameters can be set at either of two levels:

If the same parameter is set at both levels, the bot connection value takes precedence.

Add an Inworld speech provider

To add an Inworld speech provider:

  1. In the 'Authentication key' field, enter the API key from your Inworld developer console or subscription portal.

  2. Optional: In the 'TTS configuration' field, enter the advanced configuration parameters that apply to every bot connection using this provider. Inworld's own defaults apply to anything you leave out. See Configure advanced parameters at the provider level.

  3. Click Create.

    The Inworld speech provider fields

Configure advanced parameters at the provider level

The JSON you enter in 'TTS configuration' is merged into every synthesis request Live Hub sends to Inworld, so it applies to every bot connection using this provider. The parameters are Inworld's own. The full set is the Generation Request object in the Inworld API documentation.

These are the parameters most often worth setting:

Parameter What it does
applyTextNormalization Controls whether Inworld expands numbers, dates, and abbreviations before speaking them. ON or OFF.
temperature Trades consistency against expressiveness, above 0 and up to 2. Default 1.1.
maxBufferDelayMs Sets how long Inworld holds text back before it starts to synthesize it.
bufferCharThreshold Sets how many characters Inworld collects before it starts to synthesize them.

The following example stops Inworld expanding numbers and abbreviations:

{
    "applyTextNormalization": "OFF"
}

Live Hub passes on a fixed set of parameters and drops the rest, so deliveryMode has no effect.

Configure advanced parameters at the bot connection level

You can set the same parameters on a single bot connection rather than on the provider, under ttsPassthruConfig in the JSON editor on the connection's Advanced tab. See Manage bot connections.

{
    "ttsPassthruConfig": {
        "applyTextNormalization": "OFF"
    }
}

Language, model, and voice

You select the language, the model, and the voice per bot connection, on its Settings tab, not on the provider. See Manage bot connections.