In recent years, hearable technology has advanced rapidly, leading to widespread daily use in challenging acoustic environments. As their popularity has grown, so has the demand for high-quality speech communication. Although hearables can capture the users own voice with outer microphones, recordings made in noisy conditions typically require processing to enhance speech quality, which can be challenging at high noise levels. Many modern hearables also include an in-ear microphone, which is more robust to environmental noise than the outer microphones because the device partially occludes the ear canal. However, in-ear own voice recordings exhibit characteristic distortions, such as low-frequency amplification and band-limitation, which vary strongly across individuals, change during speech production, and depend on device properties. These effects need to be taken into account when using an in-ear microphone for own voice capture.The main objective of this thesis is to develop and evaluate causal deep neural network (DNN)-based own voice reconstruction (OVR) approaches that estimate clean broadband speech from noisy outer and in-ear microphone signals. Achieving this objective requires addressing several key challenges: understanding the unique distortions affecting in-ear own voice recordings, reducing the training data requirements of DNN-based OVR systems, meeting realistic computational complexity constraints, identifying suitable objective metrics for OVR performance that correlate well with subjective quality ratings, and investigating the benefits of personalizing OVR systems to individual talkers
Modern digital hearing aids go beyond basic sound amplification, offering speech enhancement, acoustic scene analysis, and wireless connectivity with other devices. Although several multi-microphone algorithms for hearing aids exist to estimate the direction of arrival (DOA) of speakers in the acoustic scene, their estimation accuracy often degrades in the presence of background noise, reverberation, and multiple simultaneously active speakers. To address this performance degradation, it has been proposed to exploit an additional external microphone (e.g., the microphone of a smartphone) that is spatially separated from the hearing devices and to jointly process all available microphone signals. While such strategies have been shown to improve DOA estimation accuracy, they typically restrict the placement of the external microphone to the vicinity of a speaker, limiting their practicality and real-world applicability. Motivated by the potential of jointly processing signals from hearing device microphones and external microphones, the primary objective of this thesis is to develop model-based binaural DOA estimation methods for multiple speakers that exploit the availability of an external microphone, without restricting its placement or requiring prior knowledge of its position. The first focus is to develop binaural-only DOA estimation methods for multiple speakers with improved robustness against noise and reverberation. The second focus is to develop strategies for exploiting an external microphone at an unknown position, i.e., not restricted to the vicinity of the speakers, in conjunction with the binaural hearing device setup, and to analyze under which conditions the external microphone provides a benefit. - www.dr.hut-verlag.de
IEEE International Conference on Acoustics, Speech and Signal Processing (50. : 2025 : Hyderabad, Indien) 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing [Piscataway, NJ] : IEEE, 2025 (2025), Seite 1-5 1 Online-Ressource
IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (2025 : Tahoe City, Calif.) Proceedings of the 2025 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics [Piscataway, NJ] : IEEE, 2025 (2025) 1 Online-Ressource
IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (2025 : Tahoe City, Calif.) Proceedings of the 2025 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics [Piscataway, NJ] : IEEE, 2025 (2025) 1 Online-Ressource
IEEE International Conference on Acoustics, Speech and Signal Processing (50. : 2025 : Hyderabad, Indien) 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing [Piscataway, NJ] : IEEE, 2025 (2025) 1 Online-Ressource
