Jul 2026· International Conference on Signal Processing and Communications· pp. 1-6· 0 citations· 20 references
Abstract
Current systems for stuttering detection are limited by task-agnostic feature compression and flat layer-fusion strategies. Traditional methods rely on Principal Component Analysis (PCA) to reduce self-supervised representations, which often suppresses subtle disfluency patterns in favor of speakeridentity cues. We propose ResGDS-HLF, a framework that replaces static compression with Residual Gumbel Dimension Selection to recover discriminative dimensions directly optimized for stuttering classification. By partitioning the transformer backbone into a functional hierarchy, i.e. acoustic, phonetic, and lexical, the model preserves specialized information that is otherwise diluted by weighted averaging. Evaluated on the SEP-28K corpus, ResGDS-HLF achieves a Macro-F1 of 0.61, representing an absolute gain of 0.14 over single-layer PCA baselines. Zero-shot transfer to FluencyBank confirms that taskdriven Gumbel selection captures robust, generalizable disfluency signatures across diverse acoustic environments.
A unified adaptation-restoration design consistently outperforms entropy-based adaptation, RAG, and perplexity-based generation over long adaptation horizons, where baselines without restoration progressively degrade while TTSD-FAR remains consistent.
Muhammad Haseeb Aslam, A. Koerich, Marco Pedersoli et al.· 0 citations
This work evaluates zero-shot phonetic classification on Chinese aspiration and Japanese moraic nasals and proposes a classification method based on continuous Articulatory Feature vectors extracted from each frame that outperforms discrete token-based methods, particularly for rare phones.
A linguistically structured approach to non-canonical phoneme recognition that decomposes phoneme prediction into articulatory feature dimensions such as manner, place, and voicing is introduced and suggests that articulatory feature supervision is a promising strategy for robust and interpretable phoneme recognition i...
Sophia Riaz, Haoze Zheng, Amos Roche et al.· 0 citations
CRAF, a cross-view residual-aware fusion framework that uses ALLM-guided cross-view attention to enrich SSL representations and adopts ALLM as a high-level reference to separate ALLM-explainable information from complementary SSL residual information, demonstrates robustness to unseen spoofing attacks.
M. Phan, Khalid Zaman, C. Mawalim et al.· 0 citations
While cascaded multi-talker ASR (MT-ASR) leverages state-of-the-art foundation models, its performance is often capped by speaker leakage during separation. Prior correction strategies primarily focus on lexical re-labeling for speaker attribution. We propose a complementary pruning-based paradigm that robustly identif...
H. Y. Nkouanga, Minwei Luo, Maggie B. Wigness et al.· 0 citations
This work refine warm-initialized LLM-based ASR models by leveraging their own pre-adaptation base LLMs by leveraging their own pre-adaptation base LLMs, focusing on LoRA-adapted settings where the base LLM is preserved.
Chan-Jan Hsu, Jaeyeon Kim, Chao-Han Huck Yang et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
