Feature Description

Overview
The Tencent Real-Time Communication (TRTC) Conversational AI solution includes built-in Speech-to-Text (STT) and interruption handling, and provides channel services, such as allowing customers to flexibly access specified AI model (Large Language Model, LLM) and Text-to-Speech (TTS) model services to create natural and smooth conversational AI features that better fit business scenarios.
Note:
The conversational AI feature is currently in beta test. To request a trial and obtain pricing information, fill out the questionnaire to contact us.
﻿
Application Scenario
Scenario
Description
Online education
In the online education scenario, real-time interaction and feedback are key to enhancing learning outcomes. By leveraging conversational AI, a platform can create a virtual teaching assistant that provides intelligent teaching assistance both in and out of class.
In class, students can ask questions to the virtual teaching assistant at any time during the teacher’s lecture to get additional explanations and better understand learning points.
Out of class, the virtual teaching assistant can provide personalized tutoring advice and learning resources based on different students' progress and needs. It can also provide responsive feedback on students' homework and questions, accompanying students in a more natural and friendly manner. Compared to lengthy textual explanations, conversational explanations can more effectively guide students and facilitate understanding.
Social entertainment
In the social entertainment scenario, conversational AI combined with real-time interaction capabilities can accurately understand users' intentions and engage in voice interactions with users, providing a more authentic and personalized social entertainment experience. A virtual companion provided by conversational AI can naturally communicate with users via voice, offering richer and more genuine emotional value compared to text. In interactive games such as Online Murder Mystery Party and Werewolf Game, conversational AI can also play the role of a host or non-player character (NPC) that engages in dynamic, real-time conversations with players and advances the storyline, allowing players to enjoy an immersive game experience.
Call center
In scenarios like online customer service, AI sales consultation, and intelligent outbound calling, conversational AI can provide a richer, real-time customer service experience. This not only effectively reduces operational costs but also significantly enhances service efficiency, offering faster service support to customers around the clock.
Highly efficient office
Through conversational AI, users can use voice to control applications, reducing manual input and making daily work easier and more efficient. Compared to text-based interactions, conversational interactions can expand the usage scenarios of various office assistants, enabling quick voice communication and task completion without being near terminal devices.
Medical assistance
By leveraging conversational AI, in scenarios such as remote diagnosis and medical consultation, patients can ask questions via voice and receive real-time personalized advice, bringing them closer to an authentic medical consultation experience. This can alleviate users' mistrust and significantly reduce their anxiety.
Note:
For details, refer to the industry practical tutorials: Emotional Companionship and Intelligent Customer Service.
Strengths
Advantage
Description
Conversational AI with ultra-low latency
In real-time AI interaction scenarios, it is crucial for LLM to promptly receive and process users' audio and video data. TRTC's ultra-low latency communication ensures that the end-to-end audio and video transmission latency is less than 300 ms globally, while keeping conversation latency below 1,000 ms. This matches human conversation response speed, allowing users to enjoy a smooth and natural interactive experience, thereby enhancing customer satisfaction.
Easy access, efficient launch
Integration can be completed in as quickly as 1 to 2 days, providing comprehensive SDK and API documentation to simplify the development process. This saves over a month of development work compared to traditional solutions, helping enterprises quickly achieve intelligent product upgrade and seize market opportunities.
Built-in accurate Automatic Speech Recognition (ASR)
The built-in advanced ASR technology supports multiple languages including English, Spanish, Japanese, Korean, Chinese and its 23 dialects, and 130 international languages. It can provide fuzzy recognition for up to 4 specified languages (excluding dialects), ensuring high accuracy and adaptability, and offering robust multilingual AI conversation support for global business.
Service model integration flexibility
Seamless integration with various LLM and TTS models is supported. We provide an integration channel that allows users to easily connect third-party LLM and TTS models. Users only need to configure the account credentials for their LLM and TTS services to seamlessly integrate them into our solution. This facilitates personalized and complex AI responses, enhancing the overall conversation experience.
Cross-platform compatibility
Multiple platforms are supported, including iOS, Android, Windows, macOS, Web, Flutter, Electron, and React Native, compatible with over 20,000 device models.
﻿

Overview
Application Scenario
Strengths

Scenario	Description
Online education	In the online education scenario, real-time interaction and feedback are key to enhancing learning outcomes. By leveraging conversational AI, a platform can create a virtual teaching assistant that provides intelligent teaching assistance both in and out of class. In class, students can ask questions to the virtual teaching assistant at any time during the teacher’s lecture to get additional explanations and better understand learning points. Out of class, the virtual teaching assistant can provide personalized tutoring advice and learning resources based on different students' progress and needs. It can also provide responsive feedback on students' homework and questions, accompanying students in a more natural and friendly manner. Compared to lengthy textual explanations, conversational explanations can more effectively guide students and facilitate understanding.
Social entertainment	In the social entertainment scenario, conversational AI combined with real-time interaction capabilities can accurately understand users' intentions and engage in voice interactions with users, providing a more authentic and personalized social entertainment experience. A virtual companion provided by conversational AI can naturally communicate with users via voice, offering richer and more genuine emotional value compared to text. In interactive games such as Online Murder Mystery Party and Werewolf Game, conversational AI can also play the role of a host or non-player character (NPC) that engages in dynamic, real-time conversations with players and advances the storyline, allowing players to enjoy an immersive game experience.
Call center	In scenarios like online customer service, AI sales consultation, and intelligent outbound calling, conversational AI can provide a richer, real-time customer service experience. This not only effectively reduces operational costs but also significantly enhances service efficiency, offering faster service support to customers around the clock.
Highly efficient office	Through conversational AI, users can use voice to control applications, reducing manual input and making daily work easier and more efficient. Compared to text-based interactions, conversational interactions can expand the usage scenarios of various office assistants, enabling quick voice communication and task completion without being near terminal devices.
Medical assistance	By leveraging conversational AI, in scenarios such as remote diagnosis and medical consultation, patients can ask questions via voice and receive real-time personalized advice, bringing them closer to an authentic medical consultation experience. This can alleviate users' mistrust and significantly reduce their anxiety.

Advantage	Description
Conversational AI with ultra-low latency	In real-time AI interaction scenarios, it is crucial for LLM to promptly receive and process users' audio and video data. TRTC's ultra-low latency communication ensures that the end-to-end audio and video transmission latency is less than 300 ms globally, while keeping conversation latency below 1,000 ms. This matches human conversation response speed, allowing users to enjoy a smooth and natural interactive experience, thereby enhancing customer satisfaction.
Easy access, efficient launch	Integration can be completed in as quickly as 1 to 2 days, providing comprehensive SDK and API documentation to simplify the development process. This saves over a month of development work compared to traditional solutions, helping enterprises quickly achieve intelligent product upgrade and seize market opportunities.
Built-in accurate Automatic Speech Recognition (ASR)	The built-in advanced ASR technology supports multiple languages including English, Spanish, Japanese, Korean, Chinese and its 23 dialects, and 130 international languages. It can provide fuzzy recognition for up to 4 specified languages (excluding dialects), ensuring high accuracy and adaptability, and offering robust multilingual AI conversation support for global business.
Service model integration flexibility	Seamless integration with various LLM and TTS models is supported. We provide an integration channel that allows users to easily connect third-party LLM and TTS models. Users only need to configure the account credentials for their LLM and TTS services to seamlessly integrate them into our solution. This facilitates personalized and complex AI responses, enhancing the overall conversation experience.
Cross-platform compatibility	Multiple platforms are supported, including iOS, Android, Windows, macOS, Web, Flutter, Electron, and React Native, compatible with over 20,000 device models.