论文解读
WaveNet-Style Guitar Amplifier Model Pruning for Real-Time iOS Deployment
音频生成 | 8.0/10
模型压缩
论文解读
Tokenizer Transplantation: Mitigating Autoregressive Collapse in Edge-Efficient Bengali ASR
语音识别 | 8.8/10
论文解读
A Quantized Native Runtime for On-Device Semantic Audio Generation
音乐生成 | 8.4/10
论文解读
It Takes Few to TANGO: A Quantized Distributed Model for Binaural Speech Enhancement
语音增强 | 6.5/10
论文解读
It Takes Few to TANGO: A Quantized Distributed Model for Binaural Speech Enhancement
语音增强 | 6.3/10
论文解读
Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs
语音识别 | 7/10
论文解读
OmniFocus: Query-Guided Modality-Balanced Token Compression for Omni-Modal Large Language Models
音视频问答 | 5.9/10
论文解读
EchoingPixels: Aliasing-Resistant Joint Token Reduction for Audio-Visual LLMs
音视频理解 | 6.3/10
论文解读
IVQ: Structured and Lightweight Vector Quantization via Binary Hierarchical Composition Inspired by \(\textit{IChing}\)
音频编码 | 8.2/10
论文解读
LightAVSeg: Lightweight Audio-Visual Segmentation
模型压缩 | 6.3/10
论文解读
OmniFit: Bridging Modalities via Layer-Adaptive Token Compression for Omnimodal Large Language Models
音视频理解 | 6.3/10
论文解读
OmniSIFT: Modality-Asymmetric Token Compression for Efficient Omni-modal Large Language Models
音视频问答 | 7.1/10
论文解读
Quaternion Self-Attention with Shared Scores
语音增强 | 6.3/10
论文解读
RT-Tango: Real-Time Distributed Binaural Speech Enhancement for Low-Power Hearing Aid Devices
语音增强 | 5.5/10
论文解读
AVTok: 1D Unified Tokenization for Holistic Audio-Video Generation
语音合成 | 6.5/10
论文解读
FlexiSLM: A Dynamic and Controllable Frame Rate Spoken Language Model
语音合成 | 7.2/10
论文解读
AVOC: Enhancing Hour-Level Audio-Video Understanding in Omni-Modal LLMs via Retrieval-Inspired Token Compression
多模态模型 | 8.4/10
论文解读
Neuromorphic Speech Enhancement with Dual-Branch Spiking Neural Networks
语音增强 | 7.0/10
论文解读
CAAD: Contrastive Audio-Aware Distillation for Efficient Speech Language Models
语音识别 | 8.9/10
论文解读
Scaling Audio Models Efficiently: A Joint Study of Compute Constraints and Optimization Behavior
语音识别 | 7.2/10