We describe a binaural auditory model for speech recognition, which is robust in the presence of reverberation and spatially separated noise intrusions. The principle underlying the model is to identify time-frequency regions which constitute reliable evidence of the speech signal. This is achieved both by determining the spatial location of the speech source, and by applying a simple model of ...

The REMOS (REverberation MOdeling for Speech recognition) concept for reverberation-robust distanttalking speech recognition [1] is presented in this paper. REMOS extends a conventional hidden Markov model (HMM) trained on close-talking data with a reverberation model describing the acoustical environment. The combination of both models is performed during recognition to match the reverberant o...

Spring-based artificial reverberation was one of the earliest attempts at compact replication of room-like reverberation for studio use. The popularity and unique sound of this effect have given it a status and desirability apart from its original use. Standard methods for modeling analog audio effects are not well suited to modeling spring reverberation, due to the complex and dispersive natur...

The auditorium within the Heydar Aliyev Center serves for an audience of 1200 and incorporates conference, concert and opera use altogether. Confronting the multi-functional performance program of the space together with uncompromising aesthetic considerations, coupled space concept has been adopted as an acoustical and architectural design aid. This paper presents mainly the dependable acousti...

This study examines the effects of background noise, actual source distance, and room reverberation on the perceived distance of a single phrase of recorded speech reproduced at a naturalistic sound pressure level. A simple rectangular room was used for stimulus generation, wherein binaural recordings were made with source-receiver distances between 0.9 m and 5.1 m, reverberation times between ...

In this paper, we propose a model to perform speech dereverberation by estimating its spectral magnitude from the reverberant counterpart. Our models are capable of extracting features that take into account both short and long-term dependencies in the signal through a convolutional encoder (which extracts features from a short, bounded context of frames) and a recurrent neural network for extr...

In everyday listening, both background noise and reverberation degrade the speech signal. While monaural speech separation based on periodicity has achieved considerable progress in handling additive noise, little research has been devoted to reverberant scenarios. Reverberation smears the harmonic structure of speech signals, and our evaluations using a pitch-based separation algorithm show th...

A series of psychoacoustic experiments were carried out to explore the relationship between an objective measure of reverberation echo density, called the normalized echo density (NED), and subjective perception of the time-domain texture of reverberation. In one experiment, 25 subjects evaluated the dissimilarity of signals having static echo densities. The reported dissimilarities matched abs...

It is well known that communication in noisy environments with reverberation present is a difficult problem to solve, particularly for the hearing-impaired listener. Two-microphone noise cancelling using an LMS adaptive filter in real time was used to process speech recorded in the presence of speech-spectrum noise at six different signal-to-noise ratios, -8, -4, 0, 4, 8, and 12 dB. Twelve norm...