Agora Marketplace Extension

Marsview Speech Analytics is a cloud-hosted or containerized API service that helps you accurately transcribe a conversation and discover insights. It is packed with models for automatic speech recognition (ASR), Intent Recognition, Tone Analysis, Natural Language Classifiers to uncover topics, keywords, entities and sentiments.
Category
Audio and video modifiers
Company
​Marsview.ai Inc.
About Us
​Marsview.ai is the leading API platform for voice and chat intelligence. Marsview APIs are used by developers to get deep insights and rich metadata from audio, video and text streams instantly with just a few lines of code. Large global brands have integrated Marsview APIs into their communication and conversational workflows to obtain greater visibility, transparency and contextual data to run efficient operations.
Marsview conversation self-service API platform offers a comprehensive suite of proprietary APIs and developer tools for automatic speech recognition, speaker separation, emotion and sentiment recognition, intent recognition, time-sequenced visual recognition and more.
Designed for the demanding call center environments that handle millions of outbound and inbound calls, Marsview APIs provide end to end workflows from call listening, recording, intent generation and voice of customer insights. Conversation APIs are also used in one-on-one to many-to-many conversations and meetings to automatically generate rich contextual feedback, key topics, moments, actions, Q&A and summaries.
Built on patent-pending advanced AI technologies, Marsview APIs are perfect for developers, product managers and business leaders to rapidly bring new products to market.
​Marsview.ai is trusted by global customers and partners, including TTEC, ServiceNow and Asurion to name a few and many other small to medium-sized businesses. Marsview.ai is headquartered in Sunnyvale, CA with an office in Bangalore. For more information, visit
Website URL: https://marsview.ai/​
Product Overview
Marsview Speech Analytics extension provides an opportunity for the entire RTE ecosystem of developers to integrate powerful AI capabilities for automatic speech recognition (ASR), speaker separation, intent recognition, tone and sentiment analysis and other conversational insights.
Developers building voice and video-driven real-time applications on Agora can use Marsview Speech Analytics API to:
Listen, record, transcribe, analyze and generate actionable insights
Understand customer’s intent and sentiments and get contextual feedback
Moderate the conversation in real-time for PII/PCI data, profanity and other unsuitable content.
Speech Analytics API works well on customer conversations, workforce meetings and outbound sales calls to automatically generate rich contextual feedback, moments, actions, Q&A, and summaries.
1
User Device With Agora SDKMarsview SystemAudio File/Voice DataTransaction IdThis transaction id  is used to  get transcribed  data from marsview system.User Device With Agora SDKMarsview System
Copied!
Introduction
Speech analytics software helps mine and analyze audio data, detecting things like emotion, tone and stress in a customer’s voice; the reason for the call; satisfaction; the products mentioned; and more. Speech analytics tools can also identify if a customer is getting upset or frustrated.
Extension workflow
If Marsview Notes extension is enabled,
 agora will create an account and project for the customer in Marsview System through a secure channel.
 After successful creation, customer can use the extension to send audio/voice data to the Marsview system. Customer will get a transaction id and is used to get transcribed data from Marsview system.
Supported Features
Convert speech into readable text from a live stream or from audio and/or video recordings, separated by speakers.
Identify the type of speech based on context and tone such as a statement, question, command and so on.
Combine the tone of voice and the sensitivity of the topic in the transcription text to classify the true sentiment as positive, negative, neutral.
Detect the type of on-screen activity as interaction, slide share, motion graphics etc. Capture key frames and slides to chapterize a visual presentation.
Generate call quality and speech quality metrics to improve conversational performance
Add custom vocabulary to generate more accurate transcriptions for your industry
Marsview’s proprietary, pre-trained AI models provide complete control, flexibility and security of your data. With just a few lines of code, developers can instantly enable Agora powered RTE applications to generate real-time insights and make rapid, data-informed decisions on the fly to be more effective, get better results, and have a greater impact on the business as a whole.
Supported Platforms:
android
Integration
Please follow the step by step process to implement Marsview Speech Analytics extension with your application
 (We are providing code samples written in Java)
 Step 1: unzip file , you will get .ar file. keep it in your native project.
 Step 2: extension package agoramarketplace.marsview.extension
Step3: you will get an api key and secret when you create a project with Marsview through agora.
Step4: you have to
 import agoramarketplace.marsview.extension.ExtensionManager;
 Step5: add the credentials to your file
1
- private final String API_KEY = "84e**** **** ****  **** 4823470a7876";
2
- private final String SECRET_KEY = "GKHJ**** **** ****-8W57GXJ";
3
- private final String USER_ID = "[email protected]";
Copied!
Step6: implement the interface io.agora.rtc2.IMediaExtensionObserver
 After enabling RTCEngine you need to follow the next step:
 Step7: Now you need to pass these variables to enable the extension
Before starting your RTC Streaming you need to enable this:
1
- mRtcEngine.enableExtension(ExtensionManager.EXTENSION_VENDOR_NAME, ExtensionManager.EXTENSION_AUDIO_FILTER_NAME, true);
Copied!
Step8: For authentication purpose, we need to pass API credentials to the extension
1
- mRtcEngine.setExtensionProperty(ExtensionManager.EXTENSION_VENDOR_NAME, ExtensionManager.EXTENSION_AUDIO_FILTER_NAME, "API_KEY", API_KEY);
Copied!
1
- mRtcEngine.setExtensionProperty(ExtensionManager.EXTENSION_VENDOR_NAME, ExtensionManager.EXTENSION_AUDIO_FILTER_NAME, "SECRET_KEY", SECRET_KEY);**
Copied!
1
- mRtcEngine.setExtensionProperty(ExtensionManager.EXTENSION_VENDOR_NAME, ExtensionManager.EXTENSION_AUDIO_FILTER_NAME, "USER_ID", USER_ID);**
Copied!
Step9: When you want to disable the transcription service, you need to set the following:
1
- mRtcEngine.enableExtension(ExtensionManager.EXTENSION_VENDOR_NAME, ExtensionManager.EXTENSION_AUDIO_FILTER_NAME, false);**
Copied!
Step10: Mainly 2 events in Marsview extension
 1. connectionState - this event describes the authentication status of the user. when the credentials do not match with the credentials he has received from agora for the extension the connection will fail
 3. transactionId - is used to fetch transcribed data from marsview
Sample code as follows:
1
- @Override
2
-  public void onEvent(String vendor, String extension, String key, String value) {
3
-        Log.d(TAG, "\nVendor: " + vendor + "\nExtension:" + extension + "\nKey:" + key + "\nValue:" + value );
4
-        if( vendor == "Marsview" && extension == "TranscriptProvider"){
5
-            if(key == "transactionId"){
6
-                // Save transaction id for future purposes
7
-            }
8
-            else if(key == "connectionState"){
9
-                try{
10
-                   JSONObject  reader = new JSONObject(value);
11
-                    String connectionState = reader.getString("connection-state");
12
-                   if(connectionState != "true"){
13
-                       // provide proper api key, sercet key , user ID
14
-                   }
15
-               }catch (Exception e){
16
-               }
17
-           }
18
-       }
19
-   }
Copied!
refer https://docs.marsview.ai/speech-analytics-api/getting-metadata​
Pricing
Plan Name
Plan Features
Pricing(USD)
Speech Analytics
1.Speech-to-Text
a.automation punctuation
b.custom vocabulary
c.sentence level keywords
(English only)
2.Spaker Separation
3. Keywords & Topics
4.Speaker Tone Analysis
5.Speaker Emotion Analysis
6.Speaker Sentiment Analysis
7.Speech/Conversation Type Detection
8.Question & Responses Detection
9.Conversation Summary
$0.05/min processed
(Example:if you consume 99.4 minutes
in a month you will be charged
100 x 0.05 = 5 USD for that month)
Useful Links
Support URL: https://docs.marsview.ai/contact-support​
​