This paper compares several particle filtering variants for speech feature enhancement in non-stationary noise environments. By analyzing the random processes of clean speech, noise and noisy speech, appropriate proposal densities are derived. The performances of the resulting particle filters, i.e. modified Sampling-ImportanceResampling (mod-SIR), auxiliary SIR and likelihood particle filter, ...

This paper compares several particle filtering variants for speech feature enhancement in non-stationary noise environments. By analyzing the random processes of clean speech, noise and noisy speech, appropriate proposal densities are derived. The performances of the resulting particle filters, i.e. modified Sampling-ImportanceResampling (mod-SIR), auxiliary SIR and likelihood particle filter, ...

The Demucs-Denoiser model has been recently shown to achieve a high level of performance for online speech enhancement, but assumes that only one source is present in the fed mixture. In real-life multiple-speech-source scenarios, it not certain which will be enhanced. To correct this issue, two target selection strategies are proposed and evaluated: (1) an embedding-based strategy, using codif...

The performance of third generation mobile systems is greatly influenced by the multiple access protocols used in the radio access system. The paper introduces a multiple access protocol, SIR (Service Integration for Radio Access), which has the potential for accommodating the requirements of speech and bursty data traffic in an efficient way. SIR is evolved from an access protocol (PRMA++) stu...

Automatic speech recognition (ASR) often fails in acoustically noisy environments. Aimed to improve speech recognition scores of an ASR in a real-life like acoustical environment, a speech pre-processing system is proposed in this paper, which consists of several stages: First, a convolutive blind source separation (BSS) is applied to the spectrogram of the signals that are pre-processed by bin...

objectives: the purpose of the present study is to investigate the oral narrative skills of late talkers mostly caused by mental disorders while they try to comprehend a wordless sequential picture story to create and narrate the relevant story. methods: to this end, 15 (10 male and 5 female) participants were who were the students of a specialized school for physically and mentally retarded st...

In this paper, a priori speech absence probability (SAP) estimator is proposed for accurately obtaining the speech presence probability (SPP) in a complex noise field. Unlike previous techniques, the proposed estimator considers a complex noise sound field where the target speech is corrupted by a coherent interference with diffuse noise around. The proposed algorithm estimates a priori SAP bas...

OBJECTIVE To assess whether speech therapy can lead to better results for early cochlear implantation (CI) children. PATIENTS A cohort of thirty-four congenitally profoundly deaf children who underwent CI before the age of 18 months at the Sixth Hospital affiliated with Shanghai Jiaotong University from January 2005 to July 2008 were included. Nineteen children received speech therapy in reha...

Spoken language acquisition is challenging for very young deaf or hard-of-hearing children (DHH) who wear hearing aids cochlear implants (CI). Timely decision-making treatment essential these and requires suitable assessments. Two such assessments are the Categories of Auditory Performance Index II (CAP-II) Speech Intelligibility Rating Scale (SIR). These have been shown to be helpful ongoing e...