An adaptive a priori SNR estimator for perceptual speech enhancement
Abstract In this paper, an adaptive averaging a priori SNR estimation employing critical band processing is proposed. The proposed method modifies the current decision-directed a priori SNR estimation to achieve faster tracking when SNR changes. The decision-directed estimator (DD) employs a fixed w...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2019-06-01
|
| Series: | EURASIP Journal on Audio, Speech, and Music Processing |
| Subjects: | |
| Online Access: | http://link.springer.com/article/10.1186/s13636-019-0150-3 |
| _version_ | 1819045849843367936 |
|---|---|
| author | Lara Nahma Pei Chee Yong Hai Huyen Dam Sven Nordholm |
| author_facet | Lara Nahma Pei Chee Yong Hai Huyen Dam Sven Nordholm |
| author_sort | Lara Nahma |
| collection | DOAJ |
| description | Abstract In this paper, an adaptive averaging a priori SNR estimation employing critical band processing is proposed. The proposed method modifies the current decision-directed a priori SNR estimation to achieve faster tracking when SNR changes. The decision-directed estimator (DD) employs a fixed weighting with the value close to one, which makes it slow in following the onsets of speech utterances. The proposed SNR estimator provides a means to solve this issue by employing an adaptive weighting factor. This allows an improved tracking of onset changes in the speech signal. As a consequence, it results in better preservation of speech components. This adaptive technique ensures that the weighting between the modified decision-directed a priori estimate and the maximum likelihood a priori estimate is a function of the speech absence probability. The estimate of the speech absence probability is modeled by a sigmoid function. Furthermore, a critical band mapping for the short-time Fourier transform analysis-synthesis system is utilized in the speech enhancement to achieve less musical noise. In addition, to evaluate the ability of the a priori SNR estimation method in preserving speech components, we proposed a modified objective measurement known as modified hamming distance. Evaluations are performed by utilizing both objective and subjective measurements. The experimental results show that the proposed method improves the speech quality under different noise conditions. Moreover, it maintains the advantage of the DD approach in eliminating the musical noise under different SNR conditions. The objective results are supported by subjective listening tests using 10 subjects (5 males and 5 females). |
| first_indexed | 2024-12-21T10:35:07Z |
| format | Article |
| id | doaj.art-9c97cdedb02d4f269181e828e0292573 |
| institution | Directory Open Access Journal |
| issn | 1687-4722 |
| language | English |
| last_indexed | 2024-12-21T10:35:07Z |
| publishDate | 2019-06-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | EURASIP Journal on Audio, Speech, and Music Processing |
| spelling | doaj.art-9c97cdedb02d4f269181e828e02925732022-12-21T19:07:05ZengSpringerOpenEURASIP Journal on Audio, Speech, and Music Processing1687-47222019-06-012019112010.1186/s13636-019-0150-3An adaptive a priori SNR estimator for perceptual speech enhancementLara Nahma0Pei Chee Yong1Hai Huyen Dam2Sven Nordholm3Department of Electrical Engineering, computing and Mathematical Sciences, Curtin UniversityNuheara Ltd.Department of Electrical Engineering, computing and Mathematical Sciences, Curtin UniversityDepartment of Electrical Engineering, computing and Mathematical Sciences, Curtin UniversityAbstract In this paper, an adaptive averaging a priori SNR estimation employing critical band processing is proposed. The proposed method modifies the current decision-directed a priori SNR estimation to achieve faster tracking when SNR changes. The decision-directed estimator (DD) employs a fixed weighting with the value close to one, which makes it slow in following the onsets of speech utterances. The proposed SNR estimator provides a means to solve this issue by employing an adaptive weighting factor. This allows an improved tracking of onset changes in the speech signal. As a consequence, it results in better preservation of speech components. This adaptive technique ensures that the weighting between the modified decision-directed a priori estimate and the maximum likelihood a priori estimate is a function of the speech absence probability. The estimate of the speech absence probability is modeled by a sigmoid function. Furthermore, a critical band mapping for the short-time Fourier transform analysis-synthesis system is utilized in the speech enhancement to achieve less musical noise. In addition, to evaluate the ability of the a priori SNR estimation method in preserving speech components, we proposed a modified objective measurement known as modified hamming distance. Evaluations are performed by utilizing both objective and subjective measurements. The experimental results show that the proposed method improves the speech quality under different noise conditions. Moreover, it maintains the advantage of the DD approach in eliminating the musical noise under different SNR conditions. The objective results are supported by subjective listening tests using 10 subjects (5 males and 5 females).http://link.springer.com/article/10.1186/s13636-019-0150-3Single-channel speech enhancementA priori SNR estimationDecision-directed approachAdaptive smoothing factorAuditory system |
| spellingShingle | Lara Nahma Pei Chee Yong Hai Huyen Dam Sven Nordholm An adaptive a priori SNR estimator for perceptual speech enhancement EURASIP Journal on Audio, Speech, and Music Processing Single-channel speech enhancement A priori SNR estimation Decision-directed approach Adaptive smoothing factor Auditory system |
| title | An adaptive a priori SNR estimator for perceptual speech enhancement |
| title_full | An adaptive a priori SNR estimator for perceptual speech enhancement |
| title_fullStr | An adaptive a priori SNR estimator for perceptual speech enhancement |
| title_full_unstemmed | An adaptive a priori SNR estimator for perceptual speech enhancement |
| title_short | An adaptive a priori SNR estimator for perceptual speech enhancement |
| title_sort | adaptive a priori snr estimator for perceptual speech enhancement |
| topic | Single-channel speech enhancement A priori SNR estimation Decision-directed approach Adaptive smoothing factor Auditory system |
| url | http://link.springer.com/article/10.1186/s13636-019-0150-3 |
| work_keys_str_mv | AT laranahma anadaptiveapriorisnrestimatorforperceptualspeechenhancement AT peicheeyong anadaptiveapriorisnrestimatorforperceptualspeechenhancement AT haihuyendam anadaptiveapriorisnrestimatorforperceptualspeechenhancement AT svennordholm anadaptiveapriorisnrestimatorforperceptualspeechenhancement AT laranahma adaptiveapriorisnrestimatorforperceptualspeechenhancement AT peicheeyong adaptiveapriorisnrestimatorforperceptualspeechenhancement AT haihuyendam adaptiveapriorisnrestimatorforperceptualspeechenhancement AT svennordholm adaptiveapriorisnrestimatorforperceptualspeechenhancement |