10 results for “streaming”
Meta unveiled Muse Voice Transcribe, a unified real-time model for streaming speech recognition, speaker diarization, and endpoint detection. The release gives developers a single API for multi-speaker, multilingual transcription at $0.18 per hour.
Google launched Gemini 3.5 Transcribe with 4.0% streaming WER and native function calling for voice agents. The model cuts transcription latency by 70% versus Chirp 3 and supports 85+ languages.
IBM and Confluent brought Granite Time Series models to Early Access on Confluent Cloud, letting teams run forecasting and anomaly detection on streaming data via Flink SQL — without a separate ML stack.
NVIDIA has released NemotronLabs VoiceChat 11B, an 11-billion-parameter end-to-end speech-to-speech model that performs streaming speech understanding. VoiceChat 11B gives researchers and infrastructure teams a reference implementation.
Google DeepMind has unveiled Gemini Robotics ER 2, a new embodied reasoning model designed to serve as a high-level brain for robots. Announced on July 30, 2026, the system introduces real-time video understanding, multi-step task.
NVIDIA will roll out DLSS 5 with 3D-Guided Neural Rendering in NBA 2K27 on September 3 at 9 p.m. Pacific Time for GeForce RTX 50 Series PCs, laptops, and GeForce NOW Ultimate. The update marks the first shipping game for a neural rendering pass that enhances lighting and materials without changing scene geometry.
ByteDance has officially launched SeedRealtime, a native audio-visual full-duplex large language model that unifies audio, video, and text within. The launch represents a significant shift in how multimodal AI systems handle real-world.
Google announced Gemini and Pixel partnerships with Arsenal, Barcelona, Bayern Munich, Liverpool, and PSG as Official Consumer AI and Smartphone partners. The deal brings AI match insights and Pixel-captured behind-the-scenes content to both men's and women's teams equally.
Alibaba launched HappyShrimp 1.0 on August 17, an end-to-end AI music model that turns natural-language prompts into complete songs with vocals and a stereo master. The release gives creators a Chinese-language-first alternative to Suno and Udio with built-in commercial licensing.
Hugging Face research shows top ASR models reproduce benchmark transcripts even when audio contradicts them. The study found models with the lowest reported error rates were most likely to regurgitate incorrect reference texts, overstating real-world performance.