ev-audio-streaming-transcription

Engage Voice Audio Streaming Quick Test Package

This package is to test basic functionality of Engage Voice Audio Streaming feature by using Google Cloud Speech-To-Text Service

Getting Started

1. Enable Google Cloud Speech-To-Text Service

• Google Cloud Account
• Enable Google Speech To Text Service
• Google Service Account

After above steps, you will get a JSON key file in your local drive. Copy the path(including file name) as {keyFilePath}.

2. Install and Use This Package

1. Create a new folder and do:

npm init

npm i ev-audio-streaming-transcription

2. Create a new file server.js with code below in it:

const { default: runServer } = require('ev-audio-streaming-transcription');

// fill in value with your {portNumber} and {keyFilePath}. eg. runServer(8000, './cred.json')
runServer({portNumber}, {keyFilePath});

And then run node server.js and you'll see:

Server started on: wss://xxxx.ngrok.io  // this is your {streamingUrl}
Client started on: https://xxxx.ngrok.io/client  // this is your {agentScriptUrl}

3. Create a streaming profile by HTTP POST to https://engage.ringcentral.com/platform/api/media/product with body:

    {
        "productType": {productType}, //QUEUE or CAMPAIGN
        "productId": {productId}, //queueId or campaignId
        "subAccountId": {subAccountId},
        "mainAccountId": {mainAccountId},
        "rcAccountId": {rcAccountId},
        "streamingUrl": {streamingUrl},
        "secret": {secret} // optional
    }

4. Create an Agent Script:

   • Log in to Engage Voice -> Admin
   • Agent Tools -> Script Designer -> (optional) Create a new group if needed
   • Add a new script -> In Queue/Campaign Assignments, tick on your Queue/Campaign -> Save
   • Open Script Studio -> From left panel, drag a Page component into the view -> Connect as from Start ---> Page ---> End
   • Hover on Page and Edit -> Add "iFrame" element and adjust the size -> Edit element and fill Resource Url with {agentScriptUrl} with ?callId={{model.call.uii}}. so https://xxxx.ngrok.io/client?callId={{model.call.uii}} -> Save

5. Now we are all set. Call the number of your Queue and transcriptions should come up.

Additional Notes

• This package only supports local debugging environment at the moment. Deployment support is WIP.