audio
30 articles about audio in AI news
OpenAI Details GPT-Live's Continuous Audio Architecture
OpenAI disclosed GPT-Live's continuous audio architecture via @rohanpaul_ai. The design shifts from turn-based voice processing to streaming, targeting latency, though specifics remain undisclosed.
ByteDance SwanTale: Unified Speech-Audio Model Hits HF
ByteDance released SwanTale, a unified speech-audio model on Hugging Face, covering voice cloning, style control, and scene synthesis. No benchmarks or technical details disclosed.
Flux 3 Beats Seedance 2.0 in BFL Tests, Adds Native Audio to Video
Black Forest Labs released Flux 3, a multimodal model generating 20-second video with native audio. Internal tests claim 52% preference over Seedance 2.0, but independent results are pending.
NVIDIA Drops 30B Nemotron Audex Audio Model with MoE
NVIDIA released Nemotron Audex 30B-A3B, a 30B-parameter MoE audio model unifying ASR, understanding, and TTS with 3B active parameters.
Wan-Streamer v0.1 Cuts Audio-Visual Interaction Latency to 200ms in Single
Wan-Streamer v0.1 achieves 200ms model-side latency in a single Transformer for full-duplex audio-visual interaction, eliminating cascaded modules. The paper lacks parameter count and benchmark comparisons, limiting reproducibility.
Gemini 3.5 Live Translate Debuts as Real-Time Audio Model
Google DeepMind released Gemini 3.5 Live Translate, an audio model for real-time translation, but disclosed no pricing, latency, or language pair details.
mlx-audio v0.4.3 Ships 6 New TTS Models, Slimmer Deps
mlx-audio v0.4.3 adds 6 TTS models, server concurrency, and slims dependencies, targeting Apple Silicon developers.
NVIDIA Nemotron 3 Nano Omni: Open Multimodal Model Unifies Video, Audio, Image, Text
NVIDIA announced Nemotron 3 Nano Omni, an open multimodal model that processes video, audio, images, and text in a unified architecture, expanding accessibility for multimodal AI research.
Pretrained Audio Models Underperform in Music Recommendation, New Research Shows
A new study evaluates nine pretrained audio models for music recommendation, finding significant performance disparity between traditional MIR tasks and both hot and cold-start recommendation scenarios.
NVIDIA's Audio Flamingo Next: 30-Min Audio, Time-Grounded Reasoning
NVIDIA has launched Audio Flamingo Next, a next-generation open audio-language model supporting 30-minute audio inputs and time-grounded reasoning. Trained on over 1 million hours of data, it reportedly outperforms larger models on key audio understanding benchmarks.
ByteDance's OmniShow Unifies Text, Image, Audio, Pose for Video Gen
ByteDance introduced OmniShow, a unified multimodal framework for video generation that accepts text, reference images, audio, and pose inputs simultaneously. It claims state-of-the-art performance across diverse conditioning settings.
PixVerse V6 Launches: 15-Second 1080P Video with Full Audio
AI video startup PixVerse launched its V6 model, capable of generating 15-second, 1080p videos with full audio from text prompts. This marks a significant upgrade in output length and quality for the platform.
Microsoft's 'Markdownify' Converts PDFs, Audio, Video to Clean LLM Markdown
Microsoft launched 'Markdownify', a Python tool that converts PDFs, Word docs, Excel, PowerPoint, audio, and YouTube URLs into clean Markdown. This addresses a major pain point in AI pipelines where raw file parsing breaks context and structure.
Qwen3.5-Omni Demonstrates 'Audio-Visual Vibe Coding' as an Emergent Ability
Alibaba's Qwen3.5-Omni model appears to have developed an emergent ability to generate code from combined audio and visual inputs without specific training. This suggests a significant leap in multimodal reasoning for a model already positioned as a strong GPT-4 competitor.
Alibaba's Qwen3.5-Omni Launches with Script-Level Captioning, Audio-Visual Vibe Coding, and Real-Time Web Search
Alibaba's Qwen team has released Qwen3.5-Omni, a multimodal model focused on interpreting images, audio, and video with new capabilities like script-level captioning and 'vibe coding'. It's open-access on Hugging Face but does not generate media.
Microsoft's VibeVoice Family Processes 60-Minute Audio in Single Pass, Eliminates Chunking for ASR & TTS
Microsoft open-sourced VibeVoice, a family of speech AI models that processes up to 60 minutes of audio without chunking. It delivers structured transcriptions with speaker diarization and generates 90-minute multi-speaker speech in one pass.
Insanely Fast Whisper CLI Transcribes 2.5 Hours of Audio in 98 Seconds with Flash Attention 2
A new open-source CLI tool called Insanely Fast Whisper achieves 19x speedup over standard Whisper large-v3, transcribing 150 minutes of audio in 98 seconds using Flash Attention 2 and batching with no quality loss.
Mistral AI Releases Voxtral TTS: 4B-Parameter Open-Weight Model Clones Voices from 3-Second Audio in 9 Languages
Mistral AI has launched Voxtral TTS, its first open-weight text-to-speech model. The 4B-parameter model clones voices from three seconds of reference audio across nine languages, with a latency of 70ms, and scored higher on naturalness than ElevenLabs Flash v2.5 in human tests.
Waves Audio Launches Lightning V3.1: 10-Second Voice Cloning with 44.1kHz Studio Quality
Waves Audio released Lightning V3.1, a voice cloning model that creates studio-quality voice replicas from just 10 seconds of audio with under 100ms latency. The update supports over 50 languages and targets real-time applications.
DEAF Benchmark Reveals Audio MLLMs Rely on Text, Not Sound, Scoring Below 50% on Acoustic Faithfulness
Researchers introduce DEAF, a 2,700-stimulus benchmark testing Audio MLLMs' acoustic processing. Evaluation of seven models shows a consistent pattern of text dominance, with models scoring below 50% on acoustic faithfulness metrics.
Fish Audio S2 Enables Word-Level Speech Control with Positional Tags, Beats GPT-4o in Human Preference Tests
Fish Audio S2 introduces a 100% open-source TTS model that uses inline positional tags for word-level vocal control, achieving 8/10 wins against GPT-4o and Gemini in human preference tests while generating audio nearly 5x faster than real-time.
OmniForcing Enables Real-Time Joint Audio-Visual Generation at 25 FPS with 0.7s Latency
Researchers introduced OmniForcing, a method that distills a bidirectional LTX-2 model into a causal streaming generator for joint audio-visual synthesis. It achieves ~25 FPS with 0.7s latency, a 35× speedup over offline diffusion models while maintaining multi-modal fidelity.
OpenAI's Bidirectional Audio Breakthrough: The End of Awkward AI Conversations
OpenAI is developing a bidirectional audio model that processes speech continuously, allowing AI to adapt instantly to interruptions. This could revolutionize voice assistants and customer support by making conversations feel truly natural.
JAEGER Breaks the 2D Barrier: How 3D Audio-Visual AI Could Transform Robotics and AR
Researchers introduce JAEGER, a framework that extends audio-visual large language models into 3D space using RGB-D and spatial audio. This breakthrough enables AI to understand and reason about physical environments with unprecedented spatial awareness.
OpenAI's Audio Revolution: New Voice Models Signal Major AI Advancements
OpenAI appears poised to release new audio models that could significantly enhance voice interaction capabilities. This development follows recent trademark filings and suggests major improvements to voice mode technology.
ByteDance Seedance 2.5 Generates 30-Second Video in One Shot
ByteDance's Seedance 2.5 generates 30-second video in one pass with 30 image, 10 video, 10 audio references. Follows Seedance 2.0 from February 2026.
Inside Shopify Hack Days: Building a prototype for music-playing pages (2026)
Shopify's 2026 Hack Days produced a prototype for music-playing product pages, involving 150 participants over 48 hours with a 200ms load time. This explores audio commerce for merchants.
Meta Trains Coding AI on Engineers' Work Traces as 8K Jobs Cut
Meta trains coding AI on engineers' work traces while cutting 8,000 jobs, per leaked audio. The behavior cloning strategy uses internal problem-solving steps as training data.
Microsoft’s VibeVoice: Open-Source Speech-to-Text with Diarization
Microsoft released VibeVoice, an MIT-licensed speech-to-text model with built-in speaker diarization. Simon Willison tested a 4-bit MLX conversion on an M5 MacBook, transcribing 1 hour of audio in ~9 minutes using ~60GB RAM.
How to Automate Meeting Notes and Action Items with Read AI's MCP Server
Integrate Read AI's MCP server with Claude Code to transform meeting audio into structured notes, decisions, and code-ready tasks without leaving your IDE.