[CfP] CS&L Special Issue on Speech Processing with Wearable and Multimodal Technologies [Deadline Dec 1st, 2026]

[Jon P Barker]

(Apologies for cross-posting)

Computer Speech & Language invites submissions for the Special Issue on Speech Processing with Wearable and Multimodal Technologies

Paper Submission Deadline: December 1st, 2026

Scope

Recent advances in wearable and multimodal technologies are creating new opportunities for speech and audio processing in realistic, everyday environments. Devices such as smart glasses, earbuds, hearing aids, headsets, mobile phones, and other body-worn sensors can provide rich combinations of audio, visual, motion, positional, and physiological signals. These technologies are particularly relevant to long-standing challenges in hands-free speech communication, including noise robustness, multi-channel processing, speech enhancement, source separation, diarization, recognition, efficient coding and transmission, and intelligibility prediction. At the same time, they raise new research questions about how to design speech processing systems that are robust, adaptive, privacy-aware, resource-efficient, and effective in unconstrained real-world conditions.

This special issue invites papers addressing speech and audio processing with wearable, multimodal, and egocentric sensing technologies. Submissions may address theoretical, algorithmic, experimental, dataset-focused, or systems-oriented aspects of the field, and may consider applications in communication, accessibility, assistive listening, healthcare, telepresence, human-computer interaction, and other real-world settings.

Topics

Topics of interest include, but are not limited to:

• Hands-free speech communication

• Microphone arrays and beamforming techniques

• Single- and multi-channel speech enhancement and separation

• Acoustic echo cancellation

• Speech recognition and robust speech representations

• Speech synthesis and voice conversion

• Speaker identification, diarization, and localization

• Voice-controlled assistants and teleconferencing systems

• Data simulation, augmentation, and adaptation of pre-trained models

• Joint end-to-end training architectures

• Robust speaker, language, and paralinguistic recognition

• Cross-environment and cross-dataset robustness

• Environmental noise modeling

• Multi-modal and ego-centric signal processing approaches

• Hearing instruments, speech quality, and intelligibility measures

• Speech technology for medical and healthcare

• Resource-constrained speech processing for wearable, mobile, and embedded communication systems

• Low-resource neural speech coding, quantization, and model compression

• Joint speech coding, enhancement, and robust transmission for real-time communication

In addition to traditional research papers, the special issue welcomes system description papers in which the main contribution lies in the design, integration, implementation, deployment, or evaluation of wearable and multimodal speech technologies. Such papers may describe successful systems, challenge entries, datasets, evaluation frameworks, or practical lessons learned from real-world deployment. For example, the recently concluded 9th CHiME Challenges provide one possible focus for contributions to the special issue. These challenges addressed conversational speech separation, speech recognition, speaker diarization, and conversational speech enhancement using wearable and multimodal sensors, including smart glasses and hearing aids. Complementing this, the recently concluded first edition of the Low-Resource Audio Codec (LRAC) challenge provides a focus on neural speech coding (including combined coding-enhancement), jointly optimised for low bitrate, low complexity, and low latency, for real-time communication on resource-constrained devices. Papers reporting results on the CHiME-9 or LRAC challenge datasets are therefore particularly encouraged. Submissions using other datasets involving real-world conversational speech, wearable sensing, multimodal input, neural speech coding and compression, resource-constrained or on-device processing, or challenging hands-free communication scenarios are also welcome.

More Details

See the following page for more details and for paper submission instructions: https://www.sciencedirect.com/special-issue/335435/speech-processing-with-wearable-and-multimodal-technologies. 

Kind Regards,

Guest Editors

Jon Barker, Christian Fuegen, Leibny Paola Garcia Perera, Shota Horiguchi, Michael Mandel, Kamil Wojcicki