Unsupervised Low Latency Speech Enhancement With RT-GCC-NMF

Wood, Sean U. N.; Rouat, Jean

doi:10.1109/jstsp.2019.2909193

Cited by 15 publications

(14 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In GCC, estimation of the GCC function

R_{x_{1}, x_{2}}

is the key step to estimate the time delay. This GCC function

R_{x_{1}, x_{2}}

is expressed mathematically in (2) [21, 34]

R_{x_{1} x_{2}} false(τ false) = \int_{- normal∞}^{\infty} ψ false(f false) G_{x_{1} x_{2} false(f false)} e^{normali 2 π f τ} normald f

where

ψ false(f false)

represents a pre‐filter and

G_{x_{1}, x_{2}} false(f false)

represents the cross spectral density (or) the spectrum of cross correlation. The cross correlation of the two time series signals can be calculated in the time domain or frequency domain.…”

Section: Time Delay Estimation (Tde)mentioning

confidence: 99%

Time delay estimation using wavelet denoising maximum likelihood method for underwater reverberant environment

Mohanty²

2020

IET Radar, Sonar & Navigation

View full text Add to dashboard Cite

“…In GCC, estimation of the GCC function

R_{x_{1}, x_{2}}

is the key step to estimate the time delay. This GCC function

R_{x_{1}, x_{2}}

is expressed mathematically in (2) [21, 34]

R_{x_{1} x_{2}} false(τ false) = \int_{- normal∞}^{\infty} ψ false(f false) G_{x_{1} x_{2} false(f false)} e^{normali 2 π f τ} normald f

where

ψ false(f false)

represents a pre‐filter and

G_{x_{1}, x_{2}} false(f false)

represents the cross spectral density (or) the spectrum of cross correlation. The cross correlation of the two time series signals can be calculated in the time domain or frequency domain.…”

Section: Time Delay Estimation (Tde)mentioning

confidence: 99%

Time delay estimation using wavelet denoising maximum likelihood method for underwater reverberant environment

Mohanty²

2020

IET Radar, Sonar & Navigation

View full text Add to dashboard Cite

“…In this study, we show that our STFT based system can also produce a comparable or better enhancement performance at an algorithmic latency as low as 4 or 2 ms. This is partially achieved by combining STFT-domain, deep learning based speech enhancement with a conventional dual window approach [36], [37], which uses a regularly long window length for STFT and a shorter window length for overlap-add. This approach is illustrated in Fig.…”

Section: Introductionmentioning

confidence: 99%

“…Our study makes three major contributions. First, we adapt a conventional dual window size approach [36], [37] to reduce the algorithmic latency of STFT-domain deep learning based speech enhancement. Second, we utilize the outputs from the first DNN for frequency-domain frame-online beamforming, and the beamforming result is fed to a second DNN for better enhancement.…”

Section: Introductionmentioning

confidence: 99%

STFT-Domain Neural Speech Enhancement with Very Low Algorithmic Latency

Wang¹,

Wichern²,

Watanabe³

et al. 2022

Preprint

View full text Add to dashboard Cite

Deep learning based speech enhancement in the short-term Fourier transform (STFT) domain typically uses a large window length such as 32 ms. A larger window contains more samples and the frequency resolution can be higher for potentially better enhancement. This however incurs an algorithmic latency of 32 ms in an online setup, because the overlap-add algorithm used in the inverse STFT (iSTFT) is also performed based on the same 32 ms window size. To reduce this inherent latency, we adapt a conventional dual window size approach, where a regular input window size is used for STFT but a shorter output window is used for the overlap-add in the iSTFT, for STFTdomain deep learning based frame-online speech enhancement. Based on this STFT and iSTFT configuration, we employ singleor multi-microphone complex spectral mapping for frame-online enhancement, where a deep neural network (DNN) is trained to predict the real and imaginary (RI) components of target speech from the mixture RI components. In addition, we use the RI components predicted by the DNN to conduct frameonline beamforming, the results of which are then used as extra features for a second DNN to perform frame-online post-filtering. The frequency-domain beamforming in between the two DNNs can be easily integrated with complex spectral mapping and is designed to not incur any algorithmic latency. Additionally, we propose a future-frame prediction technique to further reduce the algorithmic latency. Evaluation results on a noisy-reverberant speech enhancement task demonstrate the effectiveness of the proposed algorithms. Compared with Conv-TasNet, our STFTdomain system can achieve better enhancement performance for a comparable amount of computation, or comparable performance with less computation, maintaining strong performance at an algorithmic latency as low as 2 ms.

show abstract

“…In this method, a non-negative matrix factorization (NMF) dictionary is combined with generalized cross-correlation (GCC) spatial localization approach. The RT-GCC-NMF operates in a frame-by-frame manner, comparing individual dictionary atom with the desired speech signal or interfering noise based on the time-delay arrivals [30].…”

Section: Introductionmentioning

confidence: 99%

An Efficient Multistage Approach for Blind Source Separation of Noisy Convolutive Speech Mixture

Khan¹,

Jan²,

Khalil³

et al. 2021

Preprint

View full text Add to dashboard Cite

This paper proposes a novel efficient multistage algorithm to extract source speech signals from a noisy convolutive mixture. The proposed approach comprises of two stages named Blind Source Separation (BSS) and De-noising. A hybrid source prior model separates the source signals from the noisy reverberant mixture in the BSS stage. Moreover, we model the low and high-energy components by generalized multivariate Gaussian and super-Gaussian models, respectively. We use Minimum Mean Square Error (MMSE) to reduce noise in the noisy convolutive mixture signal in the de-noising stage. Furthermore, two proposed models investigate the performance gain. In the first model, the speech signal is separated from the observed noisy convolutive mixture in the BSS stage, followed by suppression of noise in the estimated source signals in the de-noising module. In the second approach, the noise is reduced using the MMSE filtering technique in the received noisy convolutive mixture at the de-noising stage, followed by separation of source signals from the de-noised reverberant mixture at the BSS stage. We evaluate the performance of the proposed scheme in terms of signal-to-distortion ratio (SDR) with respect to other well-known multistage BSS methods. The results show the superior performance of the proposed algorithm over the other state-of-the-art methods.

show abstract

Unsupervised Low Latency Speech Enhancement With RT-GCC-NMF

Cited by 15 publications

References 45 publications

Time delay estimation using wavelet denoising maximum likelihood method for underwater reverberant environment

Time delay estimation using wavelet denoising maximum likelihood method for underwater reverberant environment

STFT-Domain Neural Speech Enhancement with Very Low Algorithmic Latency

An Efficient Multistage Approach for Blind Source Separation of Noisy Convolutive Speech Mixture

Contact Info

Product

Resources

About