GitHubContribute in GitHub: Edit online

copyright: years: 2017, 2023 lastupdated: "2023-01-07"


Troubleshooting and support for Voice Gateway

When troubleshooting, keep in mind that IBM® Voice Gateway is made up of two separate required Docker containers, the SIP Orchestrator and the Media Relay, plus the optional Speech to Text Adapter container.

  • The SIP Orchestrator handles all of the SIP call signaling and the interactions with the Watson Assistant service. It also communicates with the Media Relay through a WebSocket connection, over which it sets up the media session and sends and receives text utterances.
  • The Media Relay processes the audio media streams and communicates directly with the Watson Speech to Text and Text to Speech services. The Media Relay also sources and syncs the RTP audio streams to and from the caller.
  • The Speech to Text Adapter connects to third-party speech recognition services instead of the IBM Speech to Text service.

Consider how each of these components work together as you attempt to isolate the problem. For more detail, see Architecture.

Getting help

Join our developer community to ask questions, provide answers, and get tips and tricks from the development team and other Voice Gateway users.

  • Post questions and answers in the IBM Watson Apps Community. Post your questions to the Watson Assistant discussions.
  • Open a support case from your IBM Cloud dashboard, and ask your question in the case.

If you have a licensed Voice Gateway deployment and need support, you can find additional information and open service requests from the IBM Support website.

Learn more