Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 1508.06056

A Novel Reconfigurable Hardware Design for Speech Enhancement Based on Multi-Band Spectral Subtraction Involving Magnitude and Phase Components

Tanmay Biswas·Sudhindu Bikash Mandal·Debasree Saha·Amlan Chakrabarti
View original source ↗Read source document ↗Export metadata ↓

Abstract

This paper proposes an efficient reconfigurable hardware design for speech enhancement based on multi band spectral subtraction algorithm and involving both magnitude and phase components. Our proposed design is novel as it estimates environmental noise from speech adaptively utilizing both magnitude and phase components of the speech spectrum. We performed multi-band spectrum subtraction by dividing the noisy speech spectrum into different non-uniform frequency bands having varying signal to noise ratio (SNR) and subtracting the estimated noise from each of these frequency bands. This results to the elimination of noise from both high SNR and low SNR signal components for all the frequency bands. We have coined our proposed speech enhancement technique as Multi Band Magnitude Phase Spectral Subtraction (MBMPSS). The magnitude and phase operations are executed concurrently exploiting the parallel logic blocks of Field Programmable Gate Array (FPGA), thus increasing the throughput of the system to a great extent. We have implemented our design on Spartan6 Lx45 FPGA and presented the implementation result in terms of resource utilization and delay information for the different blocks of our design. To the best of our best knowledge, this is a new type of hardware design for speech enhancement application and also a first of its kind implementation on reconfigurable hardware. We have used benchmark audio data for the evaluation of the proposed hardware and the experimental results show that our hardware shows a better SNR value compared to the existing state of the art research works.

Explore related subjects

cs.SDcs.AR

Keep this discovery

BibTeXRIS

Tanmay Biswas, Sudhindu Bikash Mandal, Debasree Saha, Amlan Chakrabarti. 2015-08-25. A Novel Reconfigurable Hardware Design for Speech Enhancement Based on Multi-Band Spectral Subtraction Involving Magnitude and Phase Components. https://arxiv.org/abs/1508.06056

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Xiaomi-CocktailASR-1 Technical Report

Recently, large language model (LLM) based ASR models have achieved significant progress, yet they generally lack support for multi-speaker scenarios, where the cocktail party problem remains a critical bottleneck for further advancing ASR. Existing TS-ASR methods, including end-to-end architectures with speaker embeddings and latest LLM-based explorations suffer from degraded single-speaker performance and the inability to reject when the target speaker is absent. In this paper, we propose Xiaomi-CocktailASR-1, an LLM-based end-to-end TS-ASR architecture. By utilizing reference speech as voiceprint prompts, it directly transcribes the target speaker's speech without requiring speech separation. Xiaomi-CocktailASR-1 maintains competitive performance in single-speaker scenarios, comparable to mainstream ASR models. It also features a negative sample rejection capability, outputting empty text when the target speaker is absent from the mixed speech. Additionally, Xiaomi-CocktailASR-1 supports a Chain-of-Thought (CoT) reasoning mode to provide explicit reasoning steps. Extensive experiments on various synthetic and real-world multispeaker benchmarks demonstrate that Xiaomi-CocktailASR-1 achieves state-of-the-art performance, effectively addressing the cocktail party problem through a unified architecture that balances multispeaker and single-speaker recognition accuracy, along with rejection capability.

cs.SD↗
arXiv2026

EConv-TasNet: Efficient Conv-TasNet for Effective Speech Separation

Conv-TasNet has served as a strong baseline for time-domain speech separation, and many studies have extended it with advanced architectures such as dual-path networks, U-Nets, and attention mechanisms. However, these methods often introduce high computational cost and complexity, limiting their deployment in resource-constrained scenarios. To address this issue, we propose eConv-TasNet, an efficient variant of Conv-TasNet that improves both effectiveness and efficiency without relying on resource-intensive modules. The proposed model consists of a group-wise early-splitting (GES) module and a multi-group feature aggregation (MGFA) module. GES generates discriminative speaker embeddings at intermediate stages, while MGFA progressively aggregates these group-level representations for refined mask estimation. Experimental results show that eConv-TasNet reduces model size by 22.4%, accelerates inference by 18.9%, and improves SI-SNRi by 14.0%-28.0% across three public benchmarks. Moreover, it achieves competitive performance compared with state-of-the-art methods while requiring significantly fewer parameters and lower inference cost. These results demonstrate a favorable efficiency-effectiveness trade-off for edge deployment.

cs.SD↗