Noise reduction by paired-microphones using spectral subtraction

Mizumachi, Mitsunori; Akagi, Masato

doi:10.1109/icassp.1998.675436

Cited by 21 publications

(15 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More sophisticated one is an adaptive beamformer [5], which controls the gain and latency parameters so as to minimize a pre-defined cost function. It is also possible to combine beamforming with spectral subtraction [6].…”

Section: Beamformingmentioning

confidence: 99%

See 1 more Smart Citation

Multi-Input Feature Combination in the Cepstral Domain for Practical Speech Recognition Systems

Obuchi

Hataoka

2009

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYIn this paper we describe a new framework of feature combination in the cepstral domain for multi-input robust speech recognition. The general framework of working in the cepstral domain has various advantages over working in the time or hypothesis domain. It is stable, easy to maintain, and less expensive because it does not require precise calibration. It is also easy to configure in a complex speech recognition system. However, it is not straightforward to improve the recognition performance by increasing the number of inputs, and we introduce the concept of variance re-scaling to compensate the negative effect of averaging several input features. Finally, we propose to take another advantage of working in the cepstral domain. The speech can be modeled using hidden Markov models, and the model can be used as prior knowledge. This approach is formulated as a new algorithm, referred to as Hypothesis-Based Feature Combination. The effectiveness of various algorithms are evaluated using two sets of speech databases. We also refer to automatic optimization of some parameters in the proposed algorithms.

show abstract

Section: Beamformingmentioning

confidence: 99%

“…Since w zero is tightly connected to all-zero feature vectors, we can expect that the recognition rate might be improved if we put the input feature vectors away from the origin by x VR = αx ave (6) where α is a scaling parameter. We call this modification variance re-scaling † † .…”

Section: Variance Re-scalingmentioning

confidence: 99%

Multi-Input Feature Combination in the Cepstral Domain for Practical Speech Recognition Systems

Obuchi

Hataoka

2009

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

show abstract

“…Those methods include noise reduction using a microphone array [6], [7], front-end noise reduction methods such as the spectral subtraction [8] or the Wiener filter [9], and model-based methods such as parallel model combination [10], [11].…”

Section: Introductionmentioning

confidence: 99%

Speech recognition in a home environment using parallel decoding with GMM-based noise modeling

Machida

Nose

Ito

2014

Signal and Information Processing Association Annual Summit and Conference (APSIPA), 2014 Asia-Pacific

View full text Add to dashboard Cite

In this paper, we propose a method for noiserobust speech recognition in a home environment based on noise modeling and parallel decoding. There are three basic ideas of the proposed method. First, we model the noise signals observed in the environment using a GMM. Second, we generate multiple noise-reduced signals using the mean vectors of the GMM and decode the signals in parallel. Third, we choose the best recognition result from the multiple recognition results based on the confidence score. The proposed method is very simple and straightforward, yet effective compared with simple noise reduction. The experiments proved that the proposed method is effective for not only noise signals in the database but also for those in the real home environment.

show abstract

“…Microphone array method is a multichannel speech enhancement method which uses a subtractive microphone array and subtracts them from the noisy speech signal using spectral subtraction (SS) [4]. There are also some methods using adaptive filtering in multichannel speech enhancement method [5,6].…”

Section: Introductionmentioning

confidence: 99%

Real-time implementation of Maximum a Posteriori (MAP) based noise reductions using Leon 3 System on Chip

Santriaji

Mauludin

Surgawiwaha

et al. 2014

2014 International Conference on Electrical Engineering and Computer Science (ICEECS)

View full text Add to dashboard Cite

Maximum a Posteriori (MAP) is an advance method to estimate noise for various audio noise reduction applications. MAP algorithm with variable speech distribution involves complex and intensive computation. Implementation in naïve method can't achieve real-time constraint. This paper proposed a method to optimize the MAP algorithm with variable speech distribution in software implementation for System on Chip (SoC). System utilizes Leon 3 microprocessor as main processing system. System is implemented in software hardware co-design to ensure flexibility and reduce computational time burden. Optimization has been done by replacing some arithmetic function with approximation function and by giving optimization option in the compiler. The simulation results show that the optimized MAP algorithm produces a linear result in SNR enhancement and faster computation time under time budget constraint.

show abstract

Noise reduction by paired-microphones using spectral subtraction

Cited by 21 publications

References 3 publications

Multi-Input Feature Combination in the Cepstral Domain for Practical Speech Recognition Systems

Multi-Input Feature Combination in the Cepstral Domain for Practical Speech Recognition Systems

Speech recognition in a home environment using parallel decoding with GMM-based noise modeling

Real-time implementation of Maximum a Posteriori (MAP) based noise reductions using Leon 3 System on Chip

Contact Info

Product

Resources

About