Neural Network-Based Artificial Bandwidth Expansion of Speech

Kontio, Jyrki; Laaksonen, Laura; Alku, Paavo

doi:10.1109/tasl.2006.885934

Cited by 43 publications

(22 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A few techniques, such as linear mapping [2], piecewise linear mapping [3,4], codebook mapping [5,6], neural networks [7,8], Gaussian mixture model [9,10], and hidden Markov model [11,12] and non-negative hidden Markov model [13], have been explored. Linear predictive coefficients (LPCs) or line spectral frequencies (LSFs) [14,15] are widely used to represent the spectral envelope, while the excitation can be found by inverse filtering the signal with LPCs, modulation techniques, non-linear processing, and the application of function generators [1].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A deep neural network approach to speech bandwidth expansion

Lee

2015

2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

108

View full text Add to dashboard Cite

We propose a deep neural network (DNN) approach to speech bandwidth expansion (BWE) by estimating the spectral mapping function from narrowband (4 kHz in bandwidth) to wideband (8 kHz in bandwidth). Log-spectrum power is used as the input and output features to perform the required nonlinear transformation, and DNNs are trained to realize this high-dimensional mapping function. When evaluating the proposed approach on a large-scale 10-hour test set, we found that the DNN-expanded speech signals give excellent objective quality measures in terms of segmental signal-to-noise ratio and log-spectral distortion when compared with conventional BWE based on Gaussian mixture models (GMMs). Subjective listening tests also give a 69% preference score for DNN-expanded speech over 31% for GMM when the phase information is assumed known. For tests in real operation when the phase information is imaged from the given narrowband signal the preference comparison goes up to 84% versus 16%. A correct phase recovery can further increase the BWE performance for the proposed DNN method.

show abstract

Section: Introductionmentioning

confidence: 99%

“…However, there are still some studies, such as folded spectrum adjusting [8] and sparse probabilistic state mapping [16]. The former one folds the narrowband spectrum and adjusts the level of the wideband spectrum, attempting to estimate the spectral envelope in a different way.…”

Section: Introductionmentioning

confidence: 99%

A deep neural network approach to speech bandwidth expansion

Lee

2015

2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

108

View full text Add to dashboard Cite

show abstract

“…The present authors have worked on ABE for several years and proposed algorithms that have been evaluated with subjective listening tests (e.g., Laaksonen et al, 2005;Kontio et al, 2007;Laaksonen et al, 2009;Pulakka et al, 2008;Pulakka and Alku, 2011) The results have been consistent and promising, indicating that ABE improves both the quality and intelligibility of narrowband speech. In addition, listening tests in three different languages (English, Mandarin Chinese, and Russian) did not show substantial differences in the ABE performance between the examined languages (Pulakka et al, 2008).…”

Section: Introductionmentioning

confidence: 55%

“…The highband envelope can be parameterized, for instance, by line-spectral frequencies (LSF) or by cepstral coefficients (Iser and Schmidt, 2005). The parameters for the highband can be estimated from narrowband signal features using, for example, codebooks (Kornagel, 2006), hidden Markov models (HMM) (Jax and Vary, 2003), Gaussian mixture models (GMM) (Park and Kim, 2000;Liu et al, 2009;Nour-Eldin and Kabal, 2011), or neural networks (Iser and Schmidt, 2005;Kontio et al, 2007).…”

Section: Introductionmentioning

confidence: 99%

Conversational quality evaluation of artificial bandwidth extension of telephone speech

Pulakka

Laaksonen

Yrttiaho

et al. 2012

The Journal of the Acoustical Society of America

Self Cite

View full text Add to dashboard Cite

Artificial bandwidth extension methods have been developed to improve the quality and intelligibility of narrowband telephone speech and to reduce the difference with wideband speech. Such methods have commonly been evaluated with objective measures or subjective listening-only tests, but conversational evaluations have been rare. This article presents a conversational evaluation of two methods for the artificial bandwidth extension of telephone speech. Bandwidth-extended narrowband speech is compared with narrowband and wideband speech in a test setting including a simulated telephone connection, realistic conversation tasks, and various background noise conditions. The responses of the subjects indicate that speech processed with one of the methods is preferred to narrowband speech in noise, but wideband speech is superior to both narrowband and bandwidth-extended speech. Bandwidth extension was found to be beneficial for telephone conversation in noisy listening conditions.

show abstract

“…For the purpose of high-quality speech synthesis, it is therefore important to compensate for the lost frequency band of the speech data. Several statistical BWE approaches have been proposed, which are based on codebooks [12], [13], Gaussian mixture models (GMMs) [14], [15], hidden Markov models (HMMs) [16], [17], neural networks [18], and deep neural networks [19], [20]. These methods basically achieve the BWE task by estimating lost high-band components from remained low-band ones.…”

Section: Introductionmentioning

confidence: 99%

Statistical Bandwidth Extension for Speech Synthesis Based on Gaussian Mixture Model with Sub-Band Basis Spectrum Model

Ohtani

Tamura

Morita

2016

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYThis paper describes a novel statistical bandwidth extension (BWE) technique based on a Gaussian mixture model (GMM) and a sub-band basis spectrum model (SBM), in which each dimensional component represents a specific acoustic space in the frequency domain. The proposed method can achieve the BWE from speech data with an arbitrary frequency bandwidth whereas the conventional methods perform the conversion from fixed narrow-band data. In the proposed method, we train a GMM with SBM parameters extracted from full-band spectra in advance. According to the bandwidth of input signal, the trained GMM is reconstructed to the GMM of the joint probability density between low-band SBM and high-band SBM components. Then high-band SBM components are estimated from low-band SBM components of the input signal based on the reconstructed GMM. Finally, BWE is achieved by adding the spectra decoded from estimated high-band SBM components to the ones of the input signal. To construct the full-band signal from the narrow-band one, we apply this method to log-amplitude spectra and aperiodic components. Objective and subjective evaluation results show that the proposed method extends the bandwidth of speech data robustly for the log-amplitude spectra. Experimental results also indicate that the aperiodic component extracted from the upsampled narrow-band signal realizes the same performance as the restored and the full-band aperiodic components in the proposed method. key words: speech enhancement, voice conversion, bandwidth extension, sub-band basis spectrum model, Gaussian mixture model

show abstract

Neural Network-Based Artificial Bandwidth Expansion of Speech

Cited by 43 publications

References 18 publications

A deep neural network approach to speech bandwidth expansion

A deep neural network approach to speech bandwidth expansion

Conversational quality evaluation of artificial bandwidth extension of telephone speech

Statistical Bandwidth Extension for Speech Synthesis Based on Gaussian Mixture Model with Sub-Band Basis Spectrum Model

Contact Info

Product

Resources

About