Microsoft has released VibeVoice-ASR as part of the VibeVoice family of open source Frontier voice AI models. VibeVoice-ASR is described as an integrated speech-to-text model that can handle up to …
Tag:
Designed
-
-
AI Tools
NVIDIA Releases PersonaPlex-7B-v1: A Real-Time Speech-to-Speech Model Designed for Natural and Full-Duplex Conversations
NVIDIA researchers released PersonaPlex-7b-v1, a full-duplex speech-to-speech conversation model that targets natural voice interactions with precise persona control. From ASR→LLM→TTS to a single full-duplex model Traditional voice assistants typically run …
-
Generative AI
Google AI releases Universal Commerce Protocol (UCP): an open-source standard designed to power the next generation of agent commerce
Can AI shopping agents go beyond sending product links and actually complete end-to-end trusted purchases inside chat? Universal Commerce Protocol, or UCP, is Google’s new open standard for agentic commerce. …
-
AI News
NVIDIA AI releases Nemotron Speech ASR: a new open source transcription model designed from the ground up for low-latency use cases like voice agents
NVIDIA recently released its new streaming English transcription model (Nemotron Speech ASR) built specifically for low-latency voice agents and live captioning. outpost nvidia/nemotron-speech-streaming-en-0.6b On Hugging Face combines a cache-aware FastConformer …
-
AI Tools
Tencent researchers release Tencent HY-MT1.5: a new translation model consisting of 1.8B and 7B models designed for seamless on-device and cloud deployment
Tencent Hunyuan researchers have released HY-MT1.5, a multilingual machine translation family that targets both mobile devices and cloud systems with the same training recipe and metrics. HY-MT1.5 includes 2 translation …
-
Generative AI
Meet LLMRouter: an intelligent routing system designed to optimize LLM inference by dynamically choosing the best-fit model for each query.
LLMRouter is an open source routing library from the U Lab at the University of Illinois Urbana-Champaign that treats model selection as a first-order system problem. It sits among a …
-
AI News
InstaDeep Introduces Nucleotide Transformer v3 (NTv3): A New Multi-Species Genomics Foundation Model, Designed for Single-Nucleotide Resolution at 1 Mb Reference Length
Genomic prediction and design now require models that connect local motifs to megabase scale regulatory context and that work across multiple organisms. Nucleotide Transformer v3, or NTv3, is InstaDeep’s new …