The results show that the draft sampling method strongly influences the optimal acceptance rate, with sampling without replacement outperforming sampling with replacement, and the theoretical upper bound of MDSD efficiency is measured for the first time.
Zheng-Mian Hu, Tong Zheng, V. Viswanathan et al.· International Conference on...· 26 citations· ⚡4
Unified audio models capable of audio understanding, audio generation and, increasingly, audio editing are proliferating rapidly. However, each capability is evaluated in isolation on specialized benchmarks and a basic question remains unanswered: can these models understand the audio they generate/edit i.e. do they ex...
SEA-LM, a Spatial Audio Understanding model, a layout-flexible spatial audio encoder trained on source localization and ego-centric voice activity detection objectives to encode First Order Ambisonics derived from variable-count, variable-position smart-glasses arrays via beamforming is presented.
Sonal Kumar, Sinan Hersek, Artem Dementyev et al.· 0 citations
This work develops ParA-LLM, a framework of 22 paralinguistic characteristics that surpasses state-of-the-art Audio LLMs like GPT-4o-Audio by 7.5% on ParA-Bench and releases ParA-Bench, a benchmark of 6,000 multiple-choice questions across speaker-speech, acoustic, and mixed categories.
Nishit Anand, Jia-Qi Su, Ke Chen et al.· Interspeech· 3 citations
D DuplexWorld introduces six worlds where voice agents are especially useful: banking, insurance, travel, healthcare and logistics, and Pathfinding, and shows that even the best voice agents leave substantial room for improvement on all 3 axes.
It is shown that steering vectors can be sparsified by up to 85-96% while retaining most performance, and that different steering methodologies agree on a subset of important dimensions, and that different steering methodologies agree on a subset of important dimensions.
Stephen Cheng, Sarah Wiegreffe, Dinesh Manocha· arXiv.org· 6 citations
VIBE is introduced, a novel text-and-video-to-music (T+V2M) generation model that leverages a depth-wise cross-layer conditioning mechanism that dynamically bridges the planning and diffusion refinement heads and a comprehensive reward modeling taxonomy, optimizing for both hard, verifiable constraints and soft, subjec...
This work presents TEMPO (Temporally-grounded Multi-task Post-training), the first unified model to handle audio, speech, and music timestamping tasks and introduces the first application of reinforcement learning to unified audio timestamping, using GRPO with verifiable temporal rewards that directly optimize the eval...
Apoorva Kulkarni, Kaousheik Jayakumar, Sreyan Ghosh et al.· 2 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.