Long-term Signal-to-noise Ratio Research Articles

ALTIS: A new algorithm for adaptive long-term SNR estimation in multi-talker babble

We introduce a real-time capable algorithm which estimates the long-term signal to noise ratio (SNR) of the speech in multi-talker babble noise. In real-time applications, long-term SNR is calculated over a sufficiently long moving frame of the noisy speech ending at the current time. The algorithm performs the real-time long-term SNR estimation by averaging “speech-likeness” values of multiple consecutive short-frames of the noisy speech which collectively form a long-frame with an adaptive length. The algorithm is calibrated to be insensitive to short-term fluctuations and transient changes in speech or noise level. However, it quickly responds to non-transient changes in long-term SNR by adjusting the duration of the long-frame on which the long-term SNR is measured. This ability is obtained by employing an event detector and adaptive frame duration. The event detector identifies non-transient changes of the long-term SNR and optimizes the duration of the long-frame accordingly. The algorithm was trained and tested for randomly generated speech samples corrupted with multi-talker babble. In addition to its ability to provide an adaptive long-term SNR estimation in a dynamic noisy situation, the evaluation results show that the algorithm outperforms the existing overall SNR estimation methods in multi-talker babble over a wide range of number of talkers and SNRs. The relatively low computational cost and the ability to update the estimated long-term SNR several times per second make this algorithm capable of operating in real-time speech processing applications.

Link Adaptation Schemes Based on Parametric Estimation of SNR Distribution Over Nakagami-$m$ Fading Channels

Link adaptation (LA) schemes maximize average spectral efficiency (ASE) using accurate channel state information (CSI) and knowledge of signal to noise ratio (SNR) distribution. In practical systems, partial CSI is available at the transmitter through feedback of modulation and coding scheme index, often known as channel quality indicator (CQI). However, parameters of SNR distribution ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${\Theta }$ </tex-math></inline-formula> ) are not known or only assumed to be known. We propose two LA schemes, for execution at the receiver and transmitter, respectively, each without prior knowledge about <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${\Theta }$ </tex-math></inline-formula> . In the receiver centric schemes, parametric estimation of SNR distribution is performed using unquantized SNR samples. In the transmitter centric scheme, we estimate <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${\Theta }$ </tex-math></inline-formula> from CQI samples using an iterative joint quantization-estimation algorithm. These parametric estimates form long-term SNR distribution, which is used to compute SNR switching thresholds. Using these thresholds, rate and power adaptation decisions are made based on instantaneous SNR. We have derived the maximum-likelihood estimator and Cramer–Rao lower bound for the proposed estimator using CQI feedback for Nakagami- <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$m$ </tex-math></inline-formula> fading channels and demonstrate that near optimal ASE can be achieved while using the proposed schemes as in an ideal scenario, where perfect knowledge about CSI and SNR distribution are available <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">a priori</i> .

Long-Term SNR Estimation of Speech Signals in Known and Unknown Channel Conditions

Many speech processing algorithms and applications rely on the explicit knowledge of signal-to-noise ratio (SNR) in their design and implementation. Estimating the SNR of a signal can enhance the performance of such technologies. We propose a novel method for estimating the long-term SNR of speech signals based on features, from which we can approximately detect regions of speech presence in a noisy signal. By measuring the energy in these regions, we create sets of energy ratios, from which we train regression models for different types of noise. If the type of noise that corrupts a signal is known, we use the corresponding regression model to estimate the SNR. When the noise is unknown, we use a deep neural network to find the “closest” regression model to estimate the SNR. Evaluations were done based on the TIMIT speech corpus, using noises from the NOISEX-92 noise database. Furthermore, we performed cross-corpora experiments by training on TIMIT and NOISEX-92 and testing on the Wall Street Journal speech corpus and DEMAND noise database. Our results show that our system provides accurate SNR estimations across different noise types, corpora, and that it outperforms other SNR estimation methods.

One-step synthesis of amino-functionalized ultrasmall near infrared-emitting persistent luminescent nanoparticles for in vitro and in vivo bioimaging.

Near infrared (NIR)-emitting persistent luminescent nanoparticles (NPLNPs) have attracted much attention in bioimaging because they can provide long-term in vivo imaging with a high signal-to-noise ratio (SNR). However, conventional NPLNPs with large particle sizes that lack modifiable surface groups suffer from many serious limitations in bioimaging. Herein, we report a one-step synthesis of amino-functionalized ZnGa2O4:Cr,Eu nanoparticles (ZGO) that have an ultrasmall size, where ethylenediamine served as the reactant to fabricate the ZGO as well as the surfactant ligand to control the nanocrystal size and form surface amino groups. The ZGO exhibited a narrow particle size distribution, a bright NIR emission and a long afterglow luminescence. In addition, due to the excellent conjugation ability of the surface amino groups, the ZGO can be easily conjugated with many bio-functional molecules, which has been successfully utilized to realize in vitro and in vivo imaging. More importantly, the ZGO achieved re-excitation imaging using 650 nm and 808 nm NIR light in situ, which is advantageous for long-term and higher SNR bioimaging.

Measuring the Long-Term SNRs of Static and Adaptive Compression Amplification Techniques for Speech in Noise

Multichannel wide-dynamic-range compression (WDRC) is a widely adopted amplification scheme in modern digital hearing aids. It attempts to provide individuals with loudness recruitment with superior speech intelligibility and greater listening comfort over a wider range of input levels. However, recent surveys have shown that compression processing (operating in the nonlinear regime) usually reduces the long-term signal-to-noise ratio (SNR). The purpose of this study was to determine the long-term SNR in an adaptive compression-ratio (CR) amplification scheme called adaptive wide-dynamic-range compression (AWDRC), and to determine whether this concept is better than static WDRC amplification at improving the long-term SNR for speech in noise. AWDRC uses the input short-term dynamic range to adjust the CR to maximize audibility and comfort. Various methods for evaluating the long-term SNR were used to observe the relationship between the CR and output SNR performance in AWDRC for seven typical audiograms, and to compare the results with those for static WDRC amplification. The results showed that the variation of the CR in AWDRC amplification can maintain the comfort and audibility of the output sound. In addition, the average long-term SNR improved by 0.1-5.5 dB for a flat hearing loss, by 0.2-3.4 dB for a reverse sloping hearing loss, by 1.4-4.8 dB for a high-frequency hearing loss, and by 0.3-5.7 dB for a mild-to-moderate-sloping high-frequency hearing loss relative to static WDRC amplification. The output long-term SNR differed significantly (p < .001) between static WDRC and AWDRC amplification. The results of this study show that AWDRC, which uses the characteristics of the input signal to adaptively adjust the CR, provides better long-term SNR performance than static WDRC amplification.

A Study of Adaptive WDRC in Hearing Aids under Noisy Conditions

Background noise poses a great challenge to the speech recognition capability of hearing-impaired patients fitted with hearing aid (HA) devices. In an HA system, a speech enhancement unit is generally employed to enhance the signal-to-noise ratio (SNR) of noisy speech in order to yield better speech understanding for HA users in noisy conditions. However, previous studies reported that a subsequent static amplification scheme, such as wide-dynamic-range compression (WDRC), may deteriorate the enhanced speech and thus decrease the speech recognition capability. This work examines the performance of a recently proposed adaptive WDRC (AWDRC) amplification scheme when used in conjunction with a speech enhancement method in HA signal processing. Experimental results demonstrate that when integrated with the same speech enhancement method, AWDRC outperforms WDRC, in terms of long-term SNRs, at several typical hearing loss conditions. The results suggest that AWDRC can be a better choice than WDRC when combining with speech enhancement to improve speech recognition capabilities for HA users in noisy conditions. Keywords: Adaptive WDRC, GMAPA, speech enhancement, hearing aids, NAL-NL1. Read more →

Better-ear glimpsing efficiency with symmetrically-placed interfering talkers

In listening tasks where a target speech signal is spatially separated from a masking voice, listeners can often gain a substantial advantage in performance by attending to the ear with the better signal-to-noise ratio (SNR). However, this better-ear strategy becomes much more complicated when a target talker located in front of the listener is masked by interfering talkers positioned at symmetric locations to the left and right of the target. When this happens, there are no long-term SNR advantages at either ear and the only binaural SNR advantages available are the result of complicated better-ear glimpses that vary as a function of frequency and rapidly switch back and forth between the two ears according to the natural fluctuations in the relative levels of the two masking voices. In this study, a signal processing technique was used to take the better-ear glimpses that would ordinarily be randomly distributed across the two ears in a binaural speech signal and move them all into the same ear. This resulted in a monaural signal that contained all the information available to an ideal listener using an optimal binaural glimpsing strategy. Speech intelligibility was measured with these optimized monaural stimuli and compared to performance with unprocessed binaural speech stimuli. Performance was similar in these two conditions, suggesting that listeners with normal hearing are able to efficiently extract information from better-ear glimpses that fluctuate rapidly across frequency and across the two ears.

Performance of multichannel reception with transmit antenna selection in arbitrarily distributed Nagakami fading channels

We present exact expressions for the average bit error rate (BER) and symbol error rate (SER) of different modulation techniques of a wireless system with multiple transmit and receive antennas. The receive antennas are assumed to use maximal ratio combining (MRC) or post-detection equal gain combining (EGC), whereas the transmit antenna that maximizes the output signal-to-noise ratio (SNR) is selected. Exact expressions of the moment generating function (MGF) of the output SNR and all its derivatives are also derived. We consider a Nakagami-m fading channel where the long-term SNR and fading parameters from the different transmit antennas are arbitrary and may be different from each other. For a given transmit antenna, the fading at the receive antennas is assumed to be independent and identically distributed (i.i.d). For the case when the Nakagami fading parameter m has an integer value in every channel, results are given in closed-form as a finite sum of simple terms. When fading parameters take any real value, our results are given in terms of the multivariate Lauricella hypergeometric function F <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">A</sub> <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">(n)</sup> . Numerical results for the error rates of different modulation techniques are presented.. Our results are validated by Monte Carlo simulation.

Long-Term Signal-to-Noise Ratio at the Input and Output of Amplitude-Compression Systems

We present measurements showing that the long-term signal-to-noise ratio (SNR) at the output of an amplification system that includes amplitude compression may be higher or lower than the long-term SNR at the input, dependent on interactions among the actual long-term input SNR, the modulation characteristics of the signal and noise being mixed, and the amplitude compression characteristics of the system under test. The effects demonstrated with the measurements shown here have implications for choices of test methods when comparing alternative hearing aid systems. The results of speech-recognition tests intended to compare alternative systems may be misleading or misinterpreted if the above interactions are not considered.