In September, we launched the new Nexmo Voice API, part of the Vonage API Platform. It’s an easily accessible API for developers to build voice applications in the cloud using familiar tools and languages. By using Nexmo Call Control Objects (NCCOs) based on a JSON schema, developers can build powerful voice applications including conferencing, call control and call recording with only a few lines of code.
The current functionality of the Voice API allows you to call (or receive calls from) a telephone endpoint. Our customers use this functionality extensively when building a voice proxy or an IVR type application, and telephone endpoints are a great solution for these types of use cases. But working with our developer and customer communities, we saw there was an even bigger opportunity to extend voice applications to new endpoint types such as WebSockets and SIP.
Today, we are happy to introduce support for both WebSockets and SIP endpoints, and you can register for early access here.
Connecting Nexmo’s Voice API to WebSockets allows our customers to connect PSTN calls to WebSocket endpoints. This means that any application that can host a WebSocket server can now be a participant in a Nexmo voice conversation. This enables some innovative use cases, including:
- Artificial intelligence engines and bots can be conferenced into a meeting to enable faster decision making (read about our collaboration with IBM Watson below).
- Analysis engines can now be easily integrated into voice calls to determine sentiment. For example, in a call center application, a supervisor can be automatically alerted when the sentiment engine determines that a customer is upset. The supervisor can even join the call and provide direction to the agent.
- Bots can make outbound calls to accomplish simple tasks such as making a restaurant reservation or more complex ones, such as requesting information from field experts that they can incorporate into their own learning and decision making.
- Bots can take inbound calls and make their expertise more readily available. For example, a doctor in a small village in Tanzania can call a medical expert bot and get access to the same medical advice available to specialists at the Mayo Clinic.
- Third party voice recognition, recording, and transcription engines can now be easily integrated with voice applications.
We also added direct support for SIP endpoints such as VoIP apps, devices and servers. You’ll now be able to mix and match SIP and PSTN endpoints seamlessly. This will enable new use cases such as:
- Dial a SIP endpoint into a conference call with other PSTN participants through a REST call.
- Forward incoming PSTN calls to SIP endpoints and perform all the actions you can normally perform on a PSTN call (e.g. play TTS, stream a file or record the call).
Integrating Vonage’s Nexmo Voice API with IBM Watson’s Project Intu
We’ve been working with the Watson team at IBM on an integration that extends the capabilities of Intu, an IBM cognitive platform, to the PSTN network. With this integration, Watson’s capabilities can be enhanced to reach billions of people who have access to a telephone. Later today at the Watson Developer Conference, developers will participate in an exercise demonstrating an Intu integration with the Vonage Nexmo Voice API over WebSockets, enabling a natural language voice interface to Watson over a regular telephone line. This is an example of how IBM Watson developers can now bring telephone enabled experiences to their artificial intelligence applications.
So what does this mean? You can now build powerful new integrations to marry voice with artificial intelligence to enable game-changing applications. We look forward to helping you innovate with IBM Watson’s Intu and the Nexmo Voice API!