JHJohn H. L. Hansen

John H. L. Hansen

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


John H. L. Hansen has a total of 241 co-authors. Here are the top co-authors by number of shared publications:

  • Abhijeet Sangwan

    Electrical Engineering, University of Texas at Dallas, Richardson, Texas United States 75080

    13 shared publications
  • Abhijeet Sangwan

    Jonsson School of Eng. & Comput. Sci., CRSS: Ctr. for Robust Speech Systems; UTDallas, CRSS: Ctr. for Robust Speech Systems, The University of Texas at Dallas; 800 W Campbell Rd., Richardson, TX 75080-3021

    12 shared publications
  • Harishchandra Dubey

    The University of Texas at Dallas, USA

    11 shared publications
  • Chengzhu Yu

    Jonsson School of Eng. & Comput. Sci., CRSS: Ctr. for Robust Speech Systems; UTDallas, CRSS: Ctr. for Robust Speech Systems, The University of Texas at Dallas; 800 W Campbell Rd., Richardson, TX 75080-3021

    7 shared publications
  • Hussnain Ali

    University of Texas at Dallas, Center for Robust Speech Systems, Cochlear Implant Laboratory , 800 W. Campbell Rd, EC 33, Richardson, Texas 75080, USA

    6 shared publications
  • Hussnain Ali

    Center for Robust Speech Systems, The University of Texas at Dallas , Richardson, Texas, 75075 USA;

    6 shared publications

Latest Publications

  • Deep learning–based environmental source separation and sound enhancement: Advancements for cochlear implant and normal hearing listeners

    by Ram C. M. C. Shekar, John H. L. Hansen

    Humans perceive non-linguistic sounds (NLSs) by associating auditory events with corresponding physical sources in a complex acoustic environment. However, previous studies have shown that cochlear implant (CI) users, vs normal hearing (NH) listeners, can face more severe challenges in identifying and tracking NLS. For CI listeners, this leads to limited autonomy, environmental awareness, safety, contextual navigation and daily engagement with individuals, society, and environmental situations. In earlier work, we studied NLS classification among CI and NH listeners and proposed a NLS enhancement solution to benefit CI/NH listeners. Building on this foundation, we propose here an experimental framework to investigate competing environmental sounds or NLS perception among CI and NH listeners. We introduce a two-source mixture model featuring “target” and “interference” source characteristics and develop an experimental setup for listener evaluation in three conditions: (i) mixed-baseline, (ii) source separation (SS) using the SUccessive DOwnsampling and Resampling of Multi-Resolution Features network, and (iii) source separation with non-linguistic sound enhancement (SSE) achieved by cascading SS output with our previously developed NLS enhancement technique. CI and NH listener evaluations were based on subjective ratings and forced-choice preference test based on perceptual measures: (i) interference, (ii) audio quality, and (iii) distortion. Our study shows a statistically significant improvement in interference reduction, with CI listeners demonstrating reduction for “nature” sounds with “category-matched” interference [F(2,21) = 4.935, p = 0.0175], and NH listeners exhibiting reductions across all NLS categories, with F-values ranging from [F(2,135) = 8.481, p = 0.000 339] to [F(2,135) = 32.37, p = 3.29 × 10−12]. Pairwise forced-choice test revealed preferences for SSE-processed nature and “domestic noises” among both CI and NH listeners. Our proposed experimental framework addresses key challenges in competing environmental sound perception among CI and NH listeners: (1) evaluation of SS for interference-characterized NLS mixture, (2) evaluation of environmental sound or NLS enhancement framework to improve perceptual outcomes with speech-targeted CI processing, and (3) perceptual measures to characterize NH and CI listener experience.

  • Open Access

    Adapting Whisper for Lightweight and Efficient Automatic Speech Recognition of Children for On-device Edge Applications

    by Satwik Dutta, Shruthigna Chandupatla, John Hansen

    Reliability on cloud providers for ASR inference to support child-centered voice-based applications is becoming challenging due to regulatory and privacy challenges. Motivated by a privacy-preserving design, this study aims to develop a lightweight & efficient Whisper ASR system capable of running on a Raspberry Pi. Upon evaluation of the MyST corpus and by examining various filtering strategies to fine-tune the `tiny.en' model, a Word Error Rate (WER) of 15.9% was achieved (11.8% filtered). A low-rank compression reduces the encoder size by 0.51M with 1.26x faster inference in GPU, with 11% relative WER increase. During inference on Pi, the compressed version required ~2 GFLOPS fewer computations. The RTF for both the models ranged between [0.23-0.41] for various input audio durations. Analyzing the RAM usage and CPU temperature showed that the PI was capable of handling both the tiny models, however it was noticed that small models initiated additional overhead/thermal throttling.

  • Open Access

    Best Practices and Considerations for Child Speech Corpus Collection and Curation in Educational, Clinical, and Forensic Scenarios

    by John Hansen, Satwik Dutta, Ellen Grand

    A child's spoken ability continues to change until their adult age. Until 7-8yrs, their speech sound development and language structure evolve rapidly. This dynamic shift in their spoken communication skills and data privacy make it challenging to curate technology-ready speech corpora for children. This study aims to bridge this gap and provide researchers and practitioners with the best practices and considerations for developing such a corpus based on an intended goal. Although primarily focused on educational goals, applications of child speech data have spread across fields including clinical and forensics fields. Motivated by this goal, we describe the WHO, WHAT, WHEN, and WHERE of data collection inspired by prior collection efforts and our experience/knowledge. We also provide a guide to establish collaboration, trust, and for navigating the human subjects research protocol. This study concludes with guidelines for corpus quality check, triage, and annotation.

  • Open Access

    DeepTraverse: A Depth-First Search Inspired Network for Algorithmic Visual Understanding

    by Bin Guo, John H. L. Hansen

    Conventional vision backbones, despite their success, often construct features through a largely uniform cascade of operations, offering limited explicit pathways for adaptive, iterative refinement. This raises a compelling question: can principles from classical search algorithms instill a more algorithmic, structured, and logical processing flow within these networks, leading to representations built through more interpretable, perhaps reasoning-like decision processes? We introduce DeepTraverse, a novel vision architecture directly inspired by algorithmic search strategies, enabling it to learn features through a process of systematic elucidation and adaptive refinement distinct from conventional approaches. DeepTraverse operationalizes this via two key synergistic components: recursive exploration modules that methodically deepen feature analysis along promising representational paths with parameter sharing for efficiency, and adaptive calibration modules that dynamically adjust feature salience based on evolving global context. The resulting algorithmic interplay allows DeepTraverse to intelligently construct and refine feature patterns. Comprehensive evaluations across a diverse suite of image classification benchmarks show that DeepTraverse achieves highly competitive classification accuracy and robust feature discrimination, often outperforming conventional models with similar or larger parameter counts. Our work demonstrates that integrating such algorithmic priors provides a principled and effective strategy for building more efficient, performant, and structured vision backbones.

  • Capabilities of the CCi-MOBILE cochlear implant research platform for real-time sound coding

    by Mahan Azadpour, Juliana N. Saba, John H. L. Hansen, Mario A. Svirsky

    Developed by the Center for Robust Speech Systems at the University of Texas at Dallas, in collaboration with New York University and the University of Wisconsin—Madison, CCi-MOBILE addresses a critical challenge in optimizing cochlear implant (CI) fitting and enhancing sound coding strategies. Existing clinical CI processors and research tools often lack either the necessary computational power and flexibility or the portability required for real-world testing. CCi-MOBILE bridges this gap by enabling the implementation and evaluation of diverse real-time sound coding algorithms in both laboratory and real-world settings, including those requiring synchronized bilateral stimulation. Building upon previous publications, this paper provides new detailed discussion on parameter setting for stimulus generation with CCi-MOBILE and serves as a comprehensive resource for scientists and engineers developing novel real-time sound coding and signal processing strategies with this platform. As part of an ongoing development effort, future generations of CCi-MOBILE may offer additional functionalities beyond those described here.


  • UniPET-SPK: A Unified Framework for Parameter-Efficient Tuning of Pre-Trained Speech Models for Robust Speaker Verification

    by Mufan Sang, John H. L. Hansen

    With excellent generalization ability, self-supervised speech models have shown impressive performance on various downstream speech tasks in the pre-training and fine-tuning paradigm. However, as the size of pre-trained models grows, fine-tuning becomes practically unfeasible due to expanding computation and storage requirements, as well as the risk of overfitting. In this study, we concentrate on exploring parameter-efficient tuning (PET) methods for adapting large-scale pre-trained self-supervised speech models to the speaker verification task. Correspondingly, we propose three parameter-efficient tuning methods: (i) an adapter-tuning method, (ii) a prompt-tuning method, and (iii) a unified framework that effectively incorporates adapter-tuning and prompt-tuning with a dynamically learnable gating mechanism. First, we propose the Inner+Inter Adapter framework, which inserts two types of adapters into pre-trained models, allowing for adaptation of latent features within the intermediate Transformer layers and output embeddings from all Transformer layers, through a parallel adapter design. Second, we propose the Deep Speaker Prompting method that concatenates trainable prompt tokens into the input space of pre-trained models to guide adaptation. Lastly, we propose the UniPET-SPK, a unified framework that effectively incorporates these two alternate PET methods into a single framework with a dynamic trainable gating mechanism. The proposed UniPET-SPK learns to find the optimal mixture of PET methods to match different datasets and scenarios. We conduct a comprehensive set of experiments on several datasets to validate the effectiveness of the proposed parameter-efficient tuning methods. Experimental results on the VoxCeleb, CN-Celeb, and the 1$^{\mathrm{{st}}}$48-UTD forensic datasets demonstrate that the proposed UniPET-SPK can consistently outperform the two PET methods, fine-tuning, and other parameter-efficient tuning methods, achieving superior performance while updating only 5.4% of the parameters. We further conduct experiments on the CN-Celeb and 1$^{\mathrm{{st}}}$48-UTD datasets to demonstrate the robustness and generalization ability of the proposed methods for speaker verification in different languages and more challenging scenarios.


  • Open Access

    Multi-objective non-intrusive hearing-aid speech assessment model

    by Hsin-Tien Chiang, Szu-Wei Fu, Hsin-Min Wang, Yu Tsao, John H. L. Hansen

    Because a reference signal is often unavailable in real-world scenarios, reference-free speech quality and intelligibility assessment models are important for many speech processing applications. Despite a great number of deep-learning models that have been applied to build non-intrusive speech assessment approaches and achieve promising performance, studies focusing on the hearing impaired (HI) subjects are limited. This paper presents HASA-Net+, a multi-objective non-intrusive hearing-aid speech assessment model, building upon our previous work, HASA-Net. HASA-Net+ improves HASA-Net in several ways: (1) inclusivity for both normal-hearing and HI listeners, (2) integration with pre-trained speech foundation models and fine-tuning techniques, (3) expansion of predictive capabilities to cover speech quality and intelligibility in diverse conditions, including noisy, denoised, reverberant, dereverberated, and vocoded speech, thereby evaluating its robustness, and (4) validation of the generalization capability using an out-of-domain dataset.