Generative AI video, automated editing, scene detection, AI dubbing, voice cloning, and upscaling are moving from research demos into production pipelines. Get the daily rundown on which AI video capabilities are actually shipping, and which ones are still just announcements.
Agora has published a guide on using `skip_patterns` within its Conversational AI platform. This feature allows users to prevent the AI's Text-to-Speech (TTS) from vocalizing structured data like code. The `skip_patterns` are used to manage how the AI processes and presents information during UI previews, tagging, and profile updates.
Agora's Conversational AI Engine enables ultra-low latency, natural AI voice interactions. The engine is designed to integrate with any Large Language Model (LLM) and any voice. This offers technology to enhance human-AI voice interactions.
Agora has announced a multilingual speech-to-text solution that provides native-level accuracy across more than 60 languages. This low-latency Automatic Speech Recognition (ASR) technology aims to reduce 'hallucinations' and is designed for real-world applications beyond English-first models.
Agora highlights its TEN VAD (Voice Activity Detection) and Turn Detection models, designed to make AI voice agents feel more natural. These models facilitate natural speech flow and context-aware pauses for real-time interaction in AI voice agents.
Agora's blog post discusses the importance of low latency in conversational AI for user adoption and retention, highlighting how response times over 300ms negatively impact user experience. The article positions Agora's technology as a solution to achieve the necessary low latency for effective AI interactions. It emphasizes that millisecond advantages are crucial to prevent user frustration and product abandonment.
Agora shares ten lessons learned from building voice AI agents, focusing on the practical challenges and solutions encountered when moving from development to real-world deployment. The article details experiences accumulated over 12 months in this field. It provides insights into the operational aspects of voice AI agent development.
The article previews Convo AI World Japan 2025, an event focusing on the future of conversational AI in Japan. Key themes include the integration of emotion, avatars, real-time voice technology, and cultural nuances in AI development.
Agora provides guidance on how to build a live voice shopping assistant using its Conversational AI platform. This solution integrates real-time audio, Large Language Model (LLM) capabilities, and transcription for product-aware responses.
The article discusses how AI-powered speech-to-text technology is evolving, highlighting applications ranging from real-time transcription and live captions to call summarization and AI-powered responses. It focuses on the transformative impact of AI in communication. The article serves as an overview of various use cases for this technology.
The article from Agora features Andrew Seagraves of Deepgram discussing the current state of speech recognition technology. It highlights that while optimized for narrow cases, speech recognition still has significant gaps in real-world applications related to data and latency. Deepgram focuses on addressing these challenges in the voice AI domain.
The article is a tutorial detailing how to integrate conversation intelligence features into an Android application. It instructs developers on using Agora extensions in conjunction with Symbl.ai for voice and video chat applications. This allows for real-time transcription and analysis of conversations within the app.
Agora's conversational AI technology, demonstrated powering robots, received the Microsoft AI Innovation Award at CES 2025. This showcases the application of AI in interactive and automated systems.
Agora provides a guide for building a transcription service into a video call web application. This service is intended to improve accessibility and user experience in video calls.
Agora published an article detailing how to build an Agora Conversational AI backend using Fastify. The guide focuses on enabling real-time voice-first AI interactions for applications.
Agora provides a step-by-step guide on how to develop a sign language recognition application using its Video SDK. The guide aims to enhance accessibility and communication through this technological application.
Agora has published a step-by-step guide detailing how to construct a real-time speech-to-text (STT) and translation web application. The guide explains how to use the company's 7.x API and Protobuf decoding to overlay live translations directly onto video streams, facilitating multilingual collaboration.
Agora provides a real-time conversational AI solution that enables the creation of AI avatars with synchronized lip movements. This solution, utilizing Agora ConvoAI and RPM, aims to make AI interactions feel more realistic than traditional chatbots with static avatars. The article focuses on the technical capability to build such a system.
The article provides a guide on how to develop an augmented reality (AR) enabled remote assistance application for Android devices. This application integrates interactive video chat capabilities to enhance the remote assistance experience. It focuses on the practical steps for building such a tool, which utilizes AR for visual guidance and live video for communication.
Agora provides a guide on building a backend for conversational AI using Express, focusing on enabling real-time voice-first AI interactions within applications. This technical tutorial outlines the steps to integrate Agora's technology for AI functionalities.
Agora published a tutorial detailing how to build a video call application that integrates live subtitles using the Agora SDK. This step-by-step guide is aimed at improving accessibility and user experience for developers.
Agora provides a guide for building a video call application that integrates AI summarization capabilities using Google's Gemini AI. The application transcribes video calls and delivers a post-call summary.
Agora provides a guide on how to integrate live translated transcriptions into video call web applications. This feature allows for real-time translation of spoken content during video calls. The guide outlines the setup process to enhance communication.
Agora published an article detailing how to integrate face filters into live streaming applications on Android using the Agora Video SDK and Banuba Face AR SDK. The guide aims to help developers add interactive elements to their streaming apps.
Agora provides a guide on building a production-ready Python backend using FastAPI and its Conversational AI Engine. This backend is designed to power voice AI functionalities within applications.
Agora provides a guide for building a conversational AI web application using Next.js, focusing on real-time voice-first AI interactions. The article is a tutorial detailing how developers can integrate Agora's technology with Next.js to create such applications.
Agora announced a collaboration with Banuba to integrate augmented reality (AR) video capabilities into the Agora Platform. This integration aims to enhance video experiences with immersive AR features.
Agora provides a recap of the AIoT 2023 event, which focused on emerging trends, achievements, challenges, and future potential within the AIoT sector. Speakers at the event discussed various aspects of the converging fields of Artificial Intelligence and Internet of Things.
The article discusses how AI-powered fan engagement is transforming fandom through the use of lifelike celebrity avatars, IP-based characters, and interactive accessories. It explores new monetization models enabled by these AI advancements. The piece focuses on the theoretical applications rather than a specific product or company launch.
The CEE 2024 event focused on AI advancements, industry case studies, and immersive technology applications. The event showcased various innovations driven by AI relevant to the streaming and entertainment industries.
Agora's Conversational AI Extension is now available on the Dify.ai Marketplace. This extension provides real-time, low-latency voice AI capabilities, enabling developers to build more interactive AI agents.
Agora launched "Agora Skills," a new offering designed to provide AI coding agents with the necessary context from Agora's platform. This aims to accelerate the development of real-time Voice AI applications, tokens, and demonstrations.
Agora provides instructions on how to integrate 3D virtual avatars into live video streams. This process utilizes MediaPipe for real-time processing and ReadyPlayerMe for avatar generation. The integration allows for augmented interaction within Agora's live streaming environment.
Agora and Pinecone have partnered to enable the integration of Retrieval Augmented Generation (RAG) into Agora's Conversational AI platform. This allows users to build Voice AI agents that can retrieve real-time context from a knowledge base to prevent AI hallucinations. The joint solution aims to enhance the accuracy and reliability of conversational AI applications.
Agora published a guide on how to integrate AI denoising capabilities into video calls using its React Native UIKit. This functionality allows developers to add a custom button to extend the UIKit for improved audio quality during video communications.
Symbl.ai provides conversation intelligence capabilities for real-time voice and video applications. These capabilities are integrated into applications built using Agora's platform.
Agora has published pricing details for its Real-Time Transcription service, which utilizes speech-to-text technology. The service offers a pay-as-you-go model and includes a free tier to get started.
Agora has published pricing details for its Real-Time Translation service. The service offers a free tier and operates on a pay-as-you-go model.
The article provides pricing information for Agora's Convo AI Device Kit. This product is designed to enable real-time generative AI interactions on edge devices.
Agora has published pricing details for its AI Noise Suppression product. The service offers a free tier before transitioning to a pay-as-you-go model for usage.
Agora offers a Speech-to-Text API designed to convert live audio into text, enabling real-time speech recognition and transcription. The API supports multilingual captions and integration with Large Language Models (LLMs) for various applications and meetings.
Agora has launched a real-time speech-to-text translation API designed to provide live multilingual captions and transcripts. This API is available for voice, video, and streaming applications, aiming to break language barriers in communication.
Agora has unveiled a Conversational AI Engine designed to enable Large Language Models (LLMs) to power lifelike voice AI. The engine features ultra-low latency, real-time interruption handling, and noise suppression, leveraging Agora's global network.
Agora has released an AI Noise Suppression extension designed to remove over 100 types of background noise during real-time audio and video calls. The extension supports cross-platform integration, enhancing clarity in live communication applications.
Agora provides infrastructure to build low-latency voice AI agents, integrating with OpenAI's Realtime API. This offering enables developers to create lifelike AI voice agents more efficiently.
Agora has launched a live benchmark for conversational AI stacks (ASR, LLM, TTS) running on its Conversational AI Engine. This benchmark provides a ranking of popular voice AI technologies. The initiative focuses on evaluating the performance of these AI components within Agora's platform.
Agora and the TEN community support the TEN Framework, an open-source framework designed for real-time, multimodal conversational AI. This framework provides tools for developing AI applications that can process and respond to real-time, multi-sensory input.
Agora's conversational AI solutions enable developers to build and deploy multimodal AI agents capable of real-time response and reaction. These solutions are designed to provide seamless integration for AI-powered interactions.
Door G Productions, a Rhode Island-based virtual production company, is leveraging AI in both pre-production and post-production workflows. The company uses AI to augment creative projects, emphasizing the enhancement of human touch. This information was discussed during a podcast episode featuring Justin Poirier and Joe Ross from Door G Productions.
Major League Pickleball (MLP) has launched an AI officiating system, developed by Owl AI, for its 2026 season in Dallas. The system utilizes existing broadcast cameras for its operation. This marks the integration of AI for officiating in professional pickleball.
OpenAI's attempt to facilitate an AI-generated animated movie titled "Critterz" has experienced setbacks, missing its intended debut at the Cannes Film Festival. The film's progress halted after OpenAI unexpectedly shuttered its Sora AI video generator, leading co-producers Chad Nelson and James Richardson to seek a new AI partner. Despite these challenges, the team aims for a first-quarter release next year, claiming significant efficiency gains through AI-driven production with a smaller crew.
