AI Briefing#audio-generation
The latest AI and developer news about #audio-generation, with the original source and a short summary.
The latest AI and developer news about #audio-generation, with the original source and a short summary.
The 4-bit version requires only 4 GB of VRAM and achieves approximately 13x real-time speed on an RTX 5090.
The new model reduces clipping on high-frequency sounds and allows users to adjust prompt influence above the default 0.3 for tighter control.
SFX v3 handles multi-sound prompts more reliably and produces cleaner audio than SFX v2 in benchmark testing.
Architecture and performance data released for serving 79 model rows, from text to robot actions, within a single runtime
Eleven v4 supports phoneme rules and inline IPA but does not support SSML phoneme tags.
StepAudio 3 Gen achieves state-of-the-art performance in TTS and multimodal audio generation using an RVQ token-based discrete generation architecture.
TontaubeV1, a 2.9B parameter open-weight TTS model based on character-level tokenization and DualCodec, has been released.
Announces the release of the Stable Audio 3.0 DAW plugin and enhanced web-based generation features.
MiniMax has released H3, an open-weight model that generates video and stereo audio simultaneously.
Scenema Audio has been released as a ComfyUI native node that runs on 8GB VRAM.
MiniMax has released the weights for its 33B H3 model, which jointly generates video and stereo audio, but Korea is excluded from the license.
Inflect-Micro-v2, a highly efficient TTS model capable of local execution with fewer than 10 million parameters, has been released.
This introduces the audio and music generation technologies that drew attention at NeurIPS 2023, along with research achievements from LG AI Research.
Introducing the multimodal generative technologies that drew attention at NeurIPS 2023 and LG AI Research's music generation interface.
WavFlow presents a method for generating high-quality audio directly in the raw waveform space without latent compression.
Stable Audio 3 has been released as a model family for variable-length audio generation and editing.
Released Stable Audio 3, a fast latent diffusion model for variable-length audio generation and editing.
ElevenCreative has released templates that combine image, video, voice, and more.
Resemble AI has released DramaBox, a voice synthesis model based on LTX 2.3.
Phosphene has released a local video/audio generation panel for Apple Silicon.
Picsart unveiled a generative CLI for the terminal, saying it supports more than 140 AI models.
Google's new TTS model precisely controls voice style using natural language audio tags.
Google has released Gemini 3.1 Flash TTS, enabling voice style control through prompts.
ElevenLabs has unveiled Templates that bundle image, video, and audio into one workflow.