Measurement of speech intelligibility based on the PESQ approach
نویسندگان
چکیده
ITU recommendation P.862, PESQ was developed for measuring speech quality in a wide variety of conditions. This paper shows that PESQ can also be used for measuring speech intelligibility. An experiment was set up to evaluate speech intelligibility in the presence of two interfering talkers at 4 SNR conditions (0, -3, -6, and -9 dB), using 100 sentences in each condition, with five normal hearing subjects. Signals were processed by four different beamforming algorithms, as used in hearing devices. Two different presentations were used, diotic and dichotic. The correlation of P.862 with the intelligibility scores, which ranged from 9 to 98%, was 0.99 for the diotic experiment and about 0.91 for the dichotic experiment.
منابع مشابه
Secure Speech Communication Using Improved OFDM Scrambler for next Generation Mobile Communication Systems
OFDM scrambling is one of the most popular techniques for secure communication. This paper proposes a new scrambling technique based on random permutation with the pseudo random binary generator to improve the performance of OFDM scrambler. To measure the intelligibility of speech, speech transmission index (STI) and common intelligibility scale (CIS) are used. The Bit Error Rate (BER) and Sign...
متن کاملWord intelligibility estimation of noise-reduced speech
It is indispensable to establish an objective test methodology for noise-reduced speech. This paper proposes a new methodology which estimates word intelligibility of the noise-reduced speech from PESQ MOS (subjective MOS estimated by the PESQ). To evaluate the effectiveness of the proposed methodology, a word intelligibility test of the noise-reduced speech was performed by using four noise re...
متن کاملAn Assessment of the Improvement Potential of Time-Frequency Masking for Speech Dereverberation
The effect of ideal time-frequency masking (ITFM) on the intelligibility of reverberated speech is tested using objective measurement, namely STI and PESQ scores. The best choice of ITFM threshold is determined for a range of reverberation times (RTs). Four existing dereverberation algorithms are also assessed. Objective test results and informal subjective listening show that IFTM provides gre...
متن کاملImproving Deep Neural Network Based Speech Enhancement in Low SNR Environments
We propose a joint framework combining speech enhancement (SE) and voice activity detection (VAD) to increase the speech intelligibility in low signal-noise-ratio (SNR) environments. Deep Neural Networks (DNN) have recently been successfully adopted as a regression model in SE. Nonetheless, the performance in harsh environments is not always satisfactory because the noise energy is often domina...
متن کاملIncorporating Auditory Masking Properties for Speech Enhancement in presence of Near-end Noise
In mobile devices, perceived speech signal degrades significantly in the presence of background noise as it reaches directly at the listener's ears. There is a need to improve the intelligibility and quality of the received speech signal in noisy environments by incorporating speech enhancement algorithms. This paper focuses on speech enhancement method including auditory masking propertie...
متن کامل