Adaptive gain control and time warp for enhanced speech intelligibility under reverberation

Petkov, Petko N.; Stylianou, Yannis

doi:10.1109/icassp.2017.7952244

Cited by 3 publications

(5 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The recognition-related algorithms on ASR include two main categories: the pre-processing algorithms and machine learning (ML)-based ASR algorithms. The pre-processing algorithms, such as dereverberation [18][19][20][21] and denoising [30], usually greatly promote the ASR system's performance. Typical pre-processing methods are beamforming methods based on microphone arrays [31].…”

Section: Related Workmentioning

confidence: 99%

“…As shown in the lower part of Figure 1, the function unit of feature extraction is in front of the recognition. Moreover, gain controls are widely integrated to realize functions, such as adjusting the signal strength [7,17,27], improving the perceptual intelligibility [8,9], optimizing the ASR performance directly [28], or by speech enhancement [18,19,39,40].…”

Section: Related Workmentioning

confidence: 99%

“…Currently, the performance of the ASR system in many human-robot interaction scenarios is unsatisfactory due to robustness limitations, and one of the critical factors is that various practical noises make it more challenging to extract the features, such as Melfrequency cepstral coefficients (MFCC) [12][13][14], log-channel energies [15], and pitch-based features [12,16]. Some common noises have been widely researched by experts in ASR, such as background noise [9,17], reverberation [18][19][20][21], squeal noise, and noises tightly related to hardware, such as thermal noises from amplifiers [22], quantizing noises from analog to digital converters (ADCs) [23], and signal quality loss caused by coding [24], compression, and transmission [25]. However, noises related to gain controls have received less attention.…”

Section: Introductionmentioning

confidence: 99%

“…However, even for a fixed gain value, different gain control strategies correspond to different noise levels, distorting the features of speech signals to different degrees. Therefore, to improve the performance of ASR systems, it is necessary to optimize the gain control strategy in ASR or ASR-related research [18,27,28].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Automatic Speech Recognition Performance Improvement for Mandarin Based on Optimizing Gain Control Strategy

Wang

Wei

Zhang

et al. 2022

Sensors

View full text Add to dashboard Cite

Automatic speech recognition (ASR) is an essential technique of human–computer interactions; gain control is a commonly used operation in ASR. However, inappropriate gain control strategies can lead to an increase in the word error rate (WER) of ASR. As there is a current lack of sufficient theoretical analyses and proof of the relationship between gain control and WER, various unconstrained gain control strategies have been adopted on realistic ASR systems, and the optimal gain control with respect to the lowest WER, is rarely achieved. A gain control strategy named maximized original signal transmission (MOST) is proposed in this study to minimize the adverse impact of gain control on ASR systems. First, by modeling the gain control strategy, the quantitative relationship between the gain control strategy and the ASR performance was established using the noise figure index. Second, through an analysis of the quantitative relationship, an optimal MOST gain control strategy with minimal performance degradation was theoretically deduced. Finally, comprehensive comparative experiments on a Mandarin dataset show that the proposed MOST gain control strategy can significantly reduce the WER of the experimental ASR system, with a 10% mean absolute WER reduction at −9 dB gain.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Automatic Speech Recognition Performance Improvement for Mandarin Based on Optimizing Gain Control Strategy

Wang

Wei

Zhang

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…The main difference was the use of slightly shorter reverberation times in the latter study ( Table 1 ). Some forms of speech slowing were also proposed by Petkov and Stylianou in [ 57 ]. Their two-stage speech modification algorithm employs adaptive gain control and so-called time warping.…”

Section: Introductionmentioning

confidence: 99%

Acoustic Sensing Analytics Applied to Speech in Reverberation Conditions

Odya

Kotus

Kurowski

et al. 2021

Sensors

View full text Add to dashboard Cite

The paper aims to discuss a case study of sensing analytics and technology in acoustics when applied to reverberation conditions. Reverberation is one of the issues that makes speech in indoor spaces challenging to understand. This problem is particularly critical in large spaces with few absorbing or diffusing surfaces. One of the natural remedies to improve speech intelligibility in such conditions may be achieved through speaking slowly. It is possible to use algorithms that reduce the rate of speech (RoS) in real time. Therefore, the study aims to find recommended values of RoS in the context of STI (speech transmission index) in different acoustic environments. In the experiments, speech intelligibility for six impulse responses recorded in spaces with different STIs is investigated using a sentence test (for the Polish language). Fifteen subjects with normal hearing participated in these tests. The results of the analytical analysis enabled us to propose a curve specifying the maximum RoS values translating into understandable speech under given acoustic conditions. This curve can be used in speech processing control technology as well as compressive reverse acoustic sensing.

show abstract

Spectral Tilt Estimation for Speech Intelligibility Enhancement Using RNN Based on All-Pole Model

Zhang

et al. 2018

MultiMedia Modeling

View full text Add to dashboard Cite

Adaptive gain control and time warp for enhanced speech intelligibility under reverberation

Cited by 3 publications

References 15 publications

Automatic Speech Recognition Performance Improvement for Mandarin Based on Optimizing Gain Control Strategy

Automatic Speech Recognition Performance Improvement for Mandarin Based on Optimizing Gain Control Strategy

Acoustic Sensing Analytics Applied to Speech in Reverberation Conditions

Spectral Tilt Estimation for Speech Intelligibility Enhancement Using RNN Based on All-Pole Model

Contact Info

Product

Resources

About