Microsoft speech provider

A Microsoft speech provider lets Live Hub use your Azure AI Speech subscription for speech-to-text (STT), text-to-speech (TTS), or both.

Microsoft'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 a Microsoft speech provider

To add a Microsoft speech provider:

  1. From the 'Azure region' drop-down list, select the region your Speech resource is in. An Azure key is valid for one region only.

  2. In the 'Authentication Key' field, enter the subscription key from that Speech resource.

  3. Optional: Under Advanced, in the 'Azure custom domain name' field, enter the name on its own, not the whole address. For https://my-domain.cognitiveservices.azure.com, enter my-domain.

    Leave it empty unless you have restricted the Speech resource to selected networks or a private endpoint. Azure accepts a custom subdomain only in that case.

  4. Optional: In the 'Provider configuration' field, enter the advanced configuration parameters that apply to every bot connection using this provider. See Configure advanced parameters at the provider level.

    The Microsoft speech provider fields

  5. Click Create.

Configure advanced parameters at the provider level

The JSON you enter in 'Provider configuration' is merged into every request Live Hub sends to Azure, so it applies to every bot connection using this provider. Use it for a private deployment or a self-signed certificate:

Parameter What it does
ttsUrl, sttUrl Point Live Hub at a private deployment instead of Azure's public endpoints.
ttsAllowSelfSignedCert, sttAllowSelfSignedCert Allow that deployment's self-signed certificate.

Enabling a self-signed certificate makes the connection to the provider less secure and open to manipulation. Use it only where the provider cannot present a properly signed certificate.

Configure advanced parameters at the bot connection level

Recognition parameters go under sttPassthruConfig in the JSON editor on the bot connection's Advanced tab. See Manage bot connections.

Each parameter is named inside setServiceProperty, with the value Azure should carry. The following example caps how long a single recognized segment can run:

{
    "sttPassthruConfig": {
        "setServiceProperty": {
            "segmentationMaximumTimeMs": {
                "value": "20000"
            }
        }
    }
}

Give the value as a string, whatever its type. Azure carries it as a URI query parameter, which you can state explicitly by adding "channel": "UriQueryParameter" alongside the value.

The names are the Azure Speech SDK's own. The full set is in QueryParameterNames.ts.

Language and voice

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