The Third Clarity Prediction Challenge (CPC3)
In the Clarity prediction challenges, we are encouraging the development of novel approaches to the intelligibility prediction of speech processed by hearing aids.
Leveraging Cochlear Implant-Based Spatial Cues for Location-Guided Target Speaker Extraction in Dynamic Environments
We train a compact deep learning model to extract a moving target speaker in the presence of interfering speech and noise, while retaining the spatial information of the target's location in the extracted speech.
Non-Intrusive Prediction of Speech Intelligibility and Listening Effort in Real-Time
This work introduces a new real-time usable and non-intrusive version of the Binaural Speech Intelligibility Model (BSIM).
Brain-controlled hearing technology: Setting targets for real-world applications
To assess the requirements of auditory attention decoding (AAD) algorithms for real-world applications of intelligent, brain-controlled hearing technology, we conducted a series of psychoacoustic experiments which quantify the effects of various AAD parameters on user experience for normal hearing and hearing impaired listeners and illustrate that AAD user experience is dependent on the properties of the listening scene as well as the listening context.
Effect Of Broadband Gain Frequency Shaping On Horizontal Localization Performance In Individuals With Normal Hearing
The study investigates the effect of channel interaction between the bandpass filters in multi-channel Hearing Aids on sound localization abilities. This was done by implementing broadband frequency shaping at low frequency channels. The study finding indicates that custom-written MATLAB code effectively preserves spatial cues, potentially due to characteristics of the filters used and high-resolution signal processing.
Using auditory texture statistics for domain-neutral removal of background sounds
Noise suppression method exploiting the stationarity of sound textures
Recovering speech intelligibility for cochlear implants in noisy and reverberant situations using multi-microphone deep learning algorithms
Coping with noise and reverberation using multi-channel speech enhancement DNN algorithms for cochlear implants
Cortical speech tracking with different lip synchronization algorithms in virtual environments
This study compares cortical speech tracking of real and virtual speakers in different conditions.
Cortical tracking of speech: Effects of intelligibility and spectral degradation
Detecting the neural correlates of speech perception under spectrally degraded listening conditions.
Having choices for enhancing voices: Target speaker extraction in noisy multi-talker environments using deep neural networks
Target speaker-informed speech enhancement approaches can enhance speech perception in noisy multi-talker environments