Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · cs/0007014

The Sound Manifesto

Michael J. O'Donnell·Ilia Bisnovatyi
View original source ↗Read source document ↗Export metadata ↓

Abstract

Computing practice today depends on visual output to drive almost all user interaction. Other senses, such as audition, may be totally neglected, or used tangentially, or used in highly restricted specialized ways. We have excellent audio rendering through D-A conversion, but we lack rich general facilities for modeling and manipulating sound comparable in quality and flexibility to graphics. We need co-ordinated research in several disciplines to improve the use of sound as an interactive information channel. Incremental and separate improvements in synthesis, analysis, speech processing, audiology, acoustics, music, etc. will not alone produce the radical progress that we seek in sonic practice. We also need to create a new central topic of study in digital audio research. The new topic will assimilate the contributions of different disciplines on a common foundation. The key central concept that we lack is sound as a general-purpose information channel. We must investigate the structure of this information channel, which is driven by the co-operative development of auditory perception and physical sound production. Particular audible encodings, such as speech and music, illuminate sonic information by example, but they are no more sufficient for a characterization than typography is sufficient for a characterization of visual information.

Explore related subjects

cs.SD

Keep this discovery

BibTeXRIS

Michael J. O'Donnell, Ilia Bisnovatyi. 2000-07-09. The Sound Manifesto. https://doi.org/10.1117/12.409214

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↗