Time-Frequency-Based Attention Cache Memory Model for Real-Time Speech Separation

2025-05-19 · Guo Chen, Kai Li, Runxuan Yang, Xiaolin Hu

Existing causal speech separation models often underperform compared to non-causal models due to difficulties in retaining historical information. To address this, we propose the Time-Frequency Attention Cache Memory (TFACM) model, which effectively captures spatio-temporal relationships through an attention mechanism and cache memory (CM) for historical information storage. In TFACM, an LSTM layer captures frequency-relative positions, while causal modeling is applied to the time dimension using local and global representations. The CM module stores past information, and the causal attention refinement (CAR) module further enhances time-based feature representations for finer granularity. Experimental results showed that TFACM achieveed comparable performance to the SOTA TF-GridNet-Causal model, with significantly lower complexity and fewer trainable parameters. For more details, visit the project page: https://cslikai.cn/TFACM/.

📄 PDF Abstract BibTeX arXiv:2505.13094

Code (0)

등록된 구현이 없습니다.

Tasks

Speech Separation

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Attention 설명 없음

Sigmoid Activation 설명 없음

Tanh Activation 설명 없음

LSTM An LSTM is a type of recurrent neural network that addresses the vanishing gradient problem in vanilla…

FreqDepthKV: Frequency-Guided Depth Sharing for Robust KV Cache Compression in Long-Context LLM Inference

2026-07-07 · Anna Córdoba, Adam Puente Tercero, Nerea Angulo Hijo, Mar Linares Tercero 외 arxiv

Long-context LLM inference is increasingly limited by the memory and bandwidth cost of KV caches, yet aggressive compression can remove the layer-specific evidence needed for retrieval and multi-step reasoning. We introd…

Question Answering Code Generation

Revisiting Multimodal KV Cache Compression: A Frequency-Domain-Guided Outlier-KV-Aware Approach

2025-11-20 · Yaoxin Yang, Peng Ye, Xudong Tan, Chongjun Tu 외 arxiv

Multimodal large language models suffer from substantial inference overhead since multimodal KV Cache grows proportionally with the visual input length. Existing multimodal KV Cache compression methods mostly rely on att…

Multi-Segment Attention: Enabling Efficient KV-Cache Management for Faster Large Language Model Serving

2026-06-01 · Chunan Shi, Yilei Chen, Yilin Chen, Xupeng Miao 외 arxiv

Large Language Model (LLM) inference relies on key-value (KV) caches to avoid redundant attention computation. While approximate KV cache retention techniques reduce memory usage by sacrificing model accuracy, lossless a…

Computational Efficiency

Time and Memory Trade-off of KV-Cache Compression in Tensor Transformer Decoding

2025-03-14 · Yifang Chen, Xiaoyu Li, YIngyu Liang, Zhenmei Shi 외

The key-value (KV) cache in the tensor version of transformers presents a significant bottleneck during inference. While previous work analyzes the fundamental space complexity barriers in standard attention mechanisms […

TTKV: Temporal-Tiered KV Cache for Long-Context LLM Inference

2026-03-27 · Gradwell Dzikanyanga, Weihao Yang, Hao Huang, Donglei Wu 외 arxiv

Key-value (KV) caching is critical for efficient inference in large language models (LLMs), yet its memory footprint scales linearly with context length, resulting in a severe scalability bottleneck. Existing approaches …