Symbl's APIs empower developers to enable:
- Real-time analysis of free-flowing discussions to automatically surface highly relevant summary discussion topics, contextual insights, suggestive action items, follow-ups, decisions, and questions.
- Voice APIs that makes it easy to add AI-powered conversation intelligence to either telephony or WebSocket interfaces.
- Conversation APIs that provide a REST interface for managing and processing your conversation data.
- Summary UI with a fully customizable and editable reference experience that indexes a searchable transcript and shows generated actionable insights, topics, timecodes, and speaker information.
Application that demonstrates Symbl's Conversational AI Adapter with Chime SDK capabilities and easy integration.
The Symbl adapter should be initialized after connecting to your Chime video conference and your access token is generated by your backend service. For instructions on how to do this please see here.
Once your Symbl Access Token is retrieved, you can set it by setting the static property ACCESS_TOKEN
on the Symbl class.
Symbl.ACCESS_TOKEN = joinInfo.Symbl.accessToken;
Once your access token is set you can call the constructor as shown below.
/**
@param {object} chime - chime instance
@property {object} configuration : {
@property {object} credentials: {
...
@property {string} attendeeId -- Client attendee id
@property {string} externalUserId
@property {string} joinToken
...
},
@property {string} meetingId -- UUID of the meeting
},
@property {string} meeting meeting name
}
@param {object} config - Symbl Configuration
@property {number} confidenceThreshold optional | default: 0.5 | 0.0 - 1.0 minimum confidence value produce valid insight
@property {string} languageCode optional - default: 'en-US' | The language code as per the BCP 47 specification
@property {boolean} insightsEnabled optional - default: true -- false if language code is not english.
@property {boolean} speechRecognition optional - default: false -- Speaker identity to use for audio in this WebSocket connection. If omitted, no speaker identification will be used for processing.
*/
const symbl = new Symbl({
configuration: {
credentials: {
attendeeId: chime.credentials.attendeeId,
externalUserId: chime.credentials.externalUserId,
},
meetingId: "acbd0689-9b84-42f7-b8b8-9bc3aa7b057a".
},
{
confidenceThreshold: 0.5,
languageCode: 'en-US',
insightsEnabled: true,
speechRecognition: true,
}
});
Realtime closed captioning can easily be added to your Chime SDK video chat application by creating a handler with 3 callback functions:
onClosedCaptioningToggled
- Will be called whenever closed captioning is toggled on or off.subtitleCreated
- Called whenever speech is first detected and a new captioning object is created.subtitleUpdated
- Called when speech is subsequently detected
The handler can be added by calling the subscribeToCaptioningEvents
function of your Symbl instance.
const captioningHandler = {
onClosedCaptioningToggled: (ccEnabled: boolean) => {
// Implement
},
subtitleCreated: (subtitle: Caption) => {
console.warn('Subtitle created', subtitle);
// Retrieve the video element that you wish to add the subtitle tracks to.
const activeVideoElement = getActiveVideoElement() as HTMLVideoElement;
if (activeVideoElement) {
subtitle.setVideoElement(activeVideoElement);
}
},
subtitleUpdated: (subtitle: Caption) => {
const activeVideoElement = getActiveVideoElement() as HTMLVideoElement;
// Check if the video element is set correctly
if (!subtitle.videoElement && activeVideoElement) {
subtitle.setVideoElement(activeVideoElement);
}
if (activeVideoElement && subtitle.videoElement !== activeVideoElement) {
console.log('Active video element changed', activeVideoElement);
subtitle.setVideoElement(activeVideoElement);
}
},
};
symbl.subscribeToCaptioningEvents(captioningHandler);
Setting the video element that subtitles will be superimposed over should be done by calling the setVideoElement
function on the Caption
class.
If your video chat application has alternating primary video tiles, this can be used to change which element is active.
Realtime insights are generated as Symbl processes the conversation in your video chat platform.
The Symbl adapter exposes a function, subscribeToInsightEvents
, that takes a handler with a function called onInsightsCreated
.
Insights are enabled by default, or by passing a property insightsEnabled
in the config
parameter of the Symbl
constructor.
new Symbl(chimeConfiguration, {insightsEnabled: true});
By creating a handler and passing it into the subscribeToInsightEvents
function, we can create a default element or use the data included in the Insight
object returned in the handlers callback.
const insightHandler = {
onInsightCreated: (insight: Insight) => {
// Creates a predesigned insight widget;
const element = insight.createElement();
// Customize any styling
element.classList.add('mx-auto');
element.style.width = '98%';
// Get container you wish to add insights to.
const insightContainer = document.getElementById('receive-insight');
// Call add on the insight object to add it to DIV
insight.add(insightContainer);
}
};
// Subscribe to realtime insight events using the handler created above
this.symbl.subscribeToInsightEvents(insightHandler);
Realtime topics are generated as Symbl processes the conversation in your video chat platform.
When a topic is detected by Symbl and the onTopicCreated
event is emitted, a Topic
object is passed to the callback function provided in the Topic
handler.
The Symbl adapter exposes a handler function, subscribeToTopicEvents
, that has a callback function onTopicCreated
.
Topics are enabled by default.
- Example:
When a topic event is emitted from the onTopicCreate
d handler, you can use the Topic
object returned and either use the createElement function to create a default element or you can use the data included in the Topic object returned in the handlers callback to create your own element, capture the data and store as a metric, etc…
const topicHandler = {
onTopicCreated: (topic: Topic) => {
//Random font color for new topic
const content = topic.phrases;
const score = topic.score;
fontSize = score * 40 + 8;
let element = topic.createElement();
element.innerText = content;
element.style.fontSize=String(fontSize)+'px'
//In case you have a Topics document you can add this element with differnt font size of topic based on the score
document.getElementById('Topics').appendChild(element);
}
};
// Subscribe to realtime tracker events using the handler created above
this.symbl.subscribeToTopicEvents(topicHandler);
Realtime trackers are generated as Symbl processes the conversation in your video chat platform.
When an Tracker is detected by Symbl and the onTrackerCreated
event is emitted, a Tracker
object is passed to the callback function provided in the Tracker
handler.
The Tracker
class holds data about the tracker generated.
The Symbl adapter exposes a handler function, subscribeToTrackerEvents
, that has a callback function onTrackerCreated
.
Trackers are enabled by adding a list of name and vocabulary pharses in the form of disctionaries to be found in a conversation.
new Symbl(chimeConfiguration, {trackers:[
{
name: "COVID-19",
vocabulary: [
"social distancing",
"cover your face with mask",
"vaccination"
]
}
],});
Subscribing to the Tracker
publisher is achieved by passing a handler to the subscribeToTrackerEvents
function of your Symbl instance.
- Example
When an tracker event is emitted from the onTrackerCreated
handler, you can use the Tracker
object returned and either use the createElement function to create a default element or you can use the data included in the Tracker
object returned in the handlers callback to create your own element, capture the data and store as a metric, etc…
const TrackerHandler = {
onTrackerCreated: (topic: Tracker) => {
const name = tracker.name;
const matches = tracker.matches;
let currentCategoryHit=0;
//Check the number of non-empty messageRefs in current tracker
for (let i = 0; i < matches.length; i++) {
if (matches[i]["messageRefs"].length > 0) {
currentCategoryHit+=1
}
}
let element = tracker.createElement();
element.innerText = name + ':' + String(currentCategoryHit);
element.style.fontSize = String(12 + currentCategoryHit)+ 'px';
//In case you have a Trackers document you can add this element with differnt
//font size of tracker based on the number of messageRefs to know how many times the tracker was foud in the converation
document.getElementById('Trackers').appendChild(element);
}
};
// Subscribe to realtime tracker events using the handler created above
this.symbl.subscribeToTrackerEvents(trackerHandler);
You must have the following installed:
-
npm 6.11 or higher
Make sure you have Node.js version 10 and higher. To add the Amazon Chime SDK for JavaScript into an existing application, install the package directly from npm:
git clone https://github.com/symblai/symbl-adapter-for-chime-demo
cd symbl-adapter-for-chime-demo
npm install
cd demos/browser
npm install
cd ../serverless
npm install
- Create an account in the Symbl Console if you don't have one already.
- After you login, you will find your appId and appSecret on the home page.
- Create a
.env
file indemos/browser
anddemos/serverless/src
that includes your appId and appSecret as shown below.
SYMBL_APP_ID=<xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx>
SYMBL_APP_SECRET=<xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx>
The App ID and App Secret are used to authenticate your session with Symbl by generating an access token. Your App ID and Secret should not be shared or posted publicly.
This demo shows how one might implement Symbl Conversational AI Adapter for Amazon Chime SDK to build meeting applications with realtime transcription and insights using Symbl's Realtime Websockets API.
To build, test, and run demos from source you will need:
- Node 10 or higher
- npm 6.11 or higher
Make sure you have built the browser application by running
cd demos/browser
npm run build
The following will create a CloudFormation stack containing a Lambda and
API Gateway deployment that runs the meetingV2
demo. Make sure the bucket and stack names are unique to AWS.
cd demos/serverless/
npm run deploy -- -r us-east-1 -b <my-bucket> -s <my-stack-name> -a meetingV2
This script will create an S3 bucket and CloudFormation stack with Lambda and API Gateway resources required to run the demo. After the script finishes, it will output a URL that can be opened in a browser.
-
Navigate to the
demos/browser
folder:cd demos/browser
-
Start the demo application:
npm run start
-
Open http://localhost:8080 in your browser.
Browser demo applications are located in the app
folder. Current demos are:
To run the Symbl Conversation AI demo application use:
npm run start
After running start
the first time, you can speed things up on subsequent iterations by using start:fast
, e.g.
npm run start:fast
If you have any questions, feel free to reach out to us at devrelations@symbl.ai or through our Community Slack or our forum.
This guide is actively developed, and we love to hear from you! Please feel free to create an issue or open a pull request with your questions, comments, suggestions and feedback. If you liked our integration guide, please star our repo!
This library is released under the Apache License