EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

An Optimized Hybrid Deep Feature Fusion Framework for Hardware-Efficient Music Retrieval

Xing Li, Yuanyuan Xu

IEEE Access · 2026 · 14 · 52427-52439

DOI: 10.1109/access.2026.3678947 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Access, 2026. Verfasst von Xing Li und 1 weiteren Autor:innen. Die Arbeit nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper introduces an optimized hybrid deep feature fusion framework designed for hardware-efficient music retrieval, specifically architected for heterogeneous AMD Xilinx FPGA platforms. To overcome the computational bottlenecks of probabilistic indexing in large-scale retrieval, the framework integrates three specialized hardware-accelerated engines, orchestrated via asynchronous ping-pong buffering and cycle-accurate AXI4-Stream scheduling to seamlessly synchronize dataflow across the PL and AIE domains: 1) a pipelined hybrid feature fusion module synthesized via Vivado HLS, which dynamically merges deep convolutional embeddings with statistical spectral descriptors to generate robust acoustic signatures, achieving a 93.8% retrieval precision with a processing latency of only 2.1 ms for 30-second audio segments; 2) an AI Engine–accelerated Variational Inference Processor that executes complex logistic normal approximations required by Correlated Topic Models (CTM) over an AXI4-Stream interconnect, optimizing covariance matrix operations to reduce BRAM utilization by 38% compared to standard architectures while sustaining a throughput of 480 audio segments per second; and 3) a DSP-optimized Topic Likelihood Estimator utilizing DSP48E2 slices to parallelize posterior probability calculations for rapid similarity matching, reducing inference latency by $5.6\times $ relative to software-based Gibbs sampling and capping total power consumption at 14 W. Validated on three benchmark music datasets (GTZAN, FMA-Small, and MagnaTagATune), the framework demonstrates a 91.5% Top-1 retrieval accuracy, outperforming baseline CNN-only approaches by 12.7% and achieving a 45% reduction in prediction variance across diverse musical textures. The system delivers $431\times $ and $20\times $ higher energy efficiency (29 mJ per inference) than equivalent high-performance CPU- and GPU-based solutions, respectively, along with a $48\times $ smaller memory footprint (8.1 MB), facilitating deployment in edge-based music discovery systems. The complete hardware-software co-design has been verified on Xilinx Zynq UltraScale+ MPSoC devices using the Vitis 2022.2 toolchain.

Zitieren

APA

Li, X., & Xu, Y. (2026). An Optimized Hybrid Deep Feature Fusion Framework for Hardware-Efficient Music Retrieval. IEEE Access, 14, 52427-52439. https://doi.org/10.1109/access.2026.3678947

Vancouver

Li X, Xu Y. An Optimized Hybrid Deep Feature Fusion Framework for Hardware-Efficient Music Retrieval. IEEE Access. 2026;14:52427-52439. doi:10.1109/access.2026.3678947

BibTeX
@article{Li2026,
  author = {Li, Xing and Xu, Yuanyuan},
  title = {An Optimized Hybrid Deep Feature Fusion Framework for Hardware-Efficient Music Retrieval},
  journal = {IEEE Access},
  year = {2026},
  volume = {14},
  pages = {52427--52439},
  doi = {10.1109/access.2026.3678947}
}
Kostenlos starten →