LiveKit
PulseAugur coverage of LiveKit — every cluster mentioning LiveKit across labs, papers, and developer communities, ranked by signal.
- uses Voxtral 3 90%
- partners with Pipecat 70%
- used by Deepgram 70%
- uses Deepgram 70%
- used by Universal-3.5 Pro Realtime 70%
- used by Retell 70%
- competes with Pipecat 60%
- uses Universal-3.5 Pro Realtime 60%
- partners with AssemblyAI 60%
- uses AssemblyAI 60%
- competes with Twilio 60%
- competes with Retell 60%
5 day(s) with sentiment data
-
Netra and Cekura platforms offer advanced voice agent testing
Two platforms, Netra and Cekura, offer solutions for testing and monitoring voice agent performance, addressing failures that text-based analysis misses. Netra excels in detailed call analysis, providing timing breakdow…
-
AssemblyAI launches $15/million char TTS API for voice agents
AssemblyAI has launched AssemblyAI-TTS-1, a new text-to-speech API designed for voice agents in live phone calls. The service offers competitive pricing at $15 per million characters and boasts low latency, with a media…
-
OpenAI launches persistent cloud agents powered by GPT-6 Astra
OpenAI has launched Dots, persistent cloud agents designed to interact with over 4,000 applications via a plugin ecosystem. These agents run on the GPT-6 Astra model, with a focus on the economic viability of always-on …
-
Mistral AI unveils Voxtral 3 speech models and voice agent SDK
Mistral AI has launched Voxtral 3, a new family of speech models. This release includes three distinct transcription models and a text-to-speech model, alongside a voice agent SDK that integrates with LiveKit. The Voxtr…
-
Mistral AI unveils Voxtral 3 speech models and voice agent SDK
Mistral AI has launched Voxtral 3, a suite of speech models including transcription and text-to-speech capabilities, alongside a voice agent SDK. The release features three transcription models, one text-to-speech model…
-
DeepSeek releases inference toolkit, PyTorch gets TPU support, Ramp builds agent harness
DeepSeek has released DeepInfer, an open-source toolkit for large-scale deep learning inference that combines distributed inference, caching, and load balancing. Google and Meta have collaborated to bring native PyTorch…
-
Mistral AI launches Voxtral 3 for voice AI conversations
Mistral AI has released Voxtral 3, a suite of tools designed for voice-based AI interactions. This release includes a speech-to-text model, a text-to-speech model, and an SDK built on LiveKit. These components work toge…
-
Voice AI agent costs: Managed vs. self-hosted platforms analyzed
Developing a custom voice AI agent can be more cost-effective than using managed platforms like Vapi or Retell, especially at higher call volumes. A voice AI call involves five distinct services: telephony, speech-to-te…
-
AI voice agent costs slashed to 2.5¢/min via custom pipeline
A company has detailed the cost breakdown of running its AI voice agent, which handles candidate interviews. By migrating from a managed voice platform to a custom LiveKit pipeline, they reduced costs from approximately…
-
DuplexJev system enables frozen LLMs to hear beyond transcripts
Researchers have introduced DuplexJev, a novel system designed to improve the efficiency of full-duplex voice agents. Instead of traditional slow decoding, DuplexJev processes Automatic Speech Recognition (ASR) encoder …
-
Google launches Gemini 3.8 Live and Extended Thinking audio models
Google has launched Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking, new advanced audio models designed to enhance voice agent capabilities and create more natural AI conversations. Gemini 3.8 Live focuses on scal…
-
AssemblyAI guides voice agent creation with Twilio and advanced streaming models
AssemblyAI has released a guide detailing how to build voice agents using their Universal-3 Pro Streaming and Universal-3.5 Pro Realtime models, in conjunction with Twilio's communication platform. The guide offers two …
-
AssemblyAI, LiveKit Discuss Real-World Voice Agent Challenges
AssemblyAI and LiveKit recently hosted a meetup in New York focusing on the challenges of building real-world voice agents. Panelists from Boardy, a founder networking platform, and Flagler Health, an AI platform for mu…
-
AssemblyAI launches Voice Agent API with flexible integration options
AssemblyAI has introduced its Voice Agent API, offering two integration lanes for voice AI stacks. The first lane, Voice Agent API, is a fully managed service handling speech-to-text, LLM routing, and text-to-speech. Th…
-
Techpotions cuts LLM API costs by 60% with AI agent architecture changes
Techpotions successfully reduced the LLM API costs for their AI Calling Agent by 60% through architectural optimizations rather than just model selection. Key strategies included implementing model routing to use cheape…
-
Meta AI launches single real-time model for voice tasks
Meta AI has introduced Muse Voice Transcribe, a novel real-time audio perception model designed to handle speech recognition, speaker diarization, and endpointing within a single system. This model, which ranks highly o…
-
AssemblyAI guides building real-time voice agents with STT-LLM-TTS architecture
AssemblyAI is detailing how to build real-time voice agents using a chained architecture that connects speech-to-text (STT), large language models (LLMs), and text-to-speech (TTS) components. The company emphasizes the …
-
AI voice tutor Bhasha Academy built with Gemini and Murf Falcon
A developer has created Bhasha Academy, an AI-powered voice tutor designed to help students in India practice English and basic math. The application supports Hinglish conversations, offers personalized learning with pe…
-
AI voice tech advances from research to real-world applications
Recent advancements in AI voice technology are highlighted across several domains, from academic research on voice cloning and detection to practical applications in contact centers and personalized learning companions.…
-
AssemblyAI enhances voice agent transcription with context carryover
AssemblyAI has introduced a new feature called Agent Context Carryover, designed to improve the accuracy of voice agent transcriptions. This feature allows the speech-to-text model to retain awareness of the ongoing con…