FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384208072> ?p ?o ?g. }

• W4384208072 endingPage "104151" @default.
• W4384208072 startingPage "104151" @default.
• W4384208072 abstract "Voice Activity Detection (VAD) is a crucial component of Speech Enhancement (SE) for accurately estimating noise, which directly affects the SE effectiveness in improving speech quality. However, conventional non-data-driven VADs often suffer from decreased accuracy at a low signal-to-noise ratio (SNR). To address this issue, a multi-feature and cosine similarity-based multi-observation VAD algorithm (mVAD) is proposed in this study. This algorithm selects noise-robust features, with Mel-frequency Cepstral Coefficients (MFCCs) as the main features, and utilizes several optimization techniques and an adaptive threshold for background noise updating. Furthermore, the soft VAD results are smoothed with an improved exponential moving average (EMA) algorithm. Besides, a shifting window is utilized to track the mean value and obtain an adaptive threshold for converting the soft results to binary ones. Experimental results indicate that mVAD can maintain high classification accuracy down to -10 dB with an increment of approximately 28% while also being computationally efficient for the CPU time (about 1/3 of statistical model-based methods). It also maintained high robustness at SNRs less than 0 dB (Δ≤2.1%). Moreover, sometimes mVAD even achieved higher accuracy levels than deep learning-based VADs. To further demonstrate the effectiveness of the proposed method, the VAD results are used as an additional feature to train and test a neural network (NN)-based SE model, enhancing the SE performance. This study proves that mVAD does not rely on prior noise knowledge, reaching the dual effect of complexity reduction and accuracy improvement for speech enhancement, making it a promising approach for robust VAD in low SNR environments." @default.
• W4384208072 created "2023-07-14" @default.
• W4384208072 creator A5033045522 @default.
• W4384208072 creator A5046357357 @default.
• W4384208072 creator A5081589298 @default.
• W4384208072 creator A5090385327 @default.
• W4384208072 date "2023-09-01" @default.
• W4384208072 modified "2023-10-14" @default.
• W4384208072 title "A robust and lightweight voice activity detection algorithm for speech enhancement at low signal-to-noise ratio" @default.
• W4384208072 cites W1957411601 @default.
• W4384208072 cites W1985242443 @default.
• W4384208072 cites W2052384514 @default.
• W4384208072 cites W2098265087 @default.
• W4384208072 cites W2100555417 @default.
• W4384208072 cites W2121973264 @default.
• W4384208072 cites W2126554103 @default.
• W4384208072 cites W2129120544 @default.
• W4384208072 cites W2137449599 @default.
• W4384208072 cites W2149053750 @default.
• W4384208072 cites W2159105055 @default.
• W4384208072 cites W2181253919 @default.
• W4384208072 cites W2194940824 @default.
• W4384208072 cites W2260826958 @default.
• W4384208072 cites W2341283081 @default.
• W4384208072 cites W2576958945 @default.
• W4384208072 cites W2793549079 @default.
• W4384208072 cites W2954930777 @default.
• W4384208072 cites W3038454187 @default.
• W4384208072 cites W3042422286 @default.
• W4384208072 cites W4281554565 @default.
• W4384208072 doi "https://doi.org/10.1016/j.dsp.2023.104151" @default.
• W4384208072 hasPublicationYear "2023" @default.
• W4384208072 type Work @default.
• W4384208072 citedByCount "0" @default.
• W4384208072 crossrefType "journal-article" @default.
• W4384208072 hasAuthorship W4384208072A5033045522 @default.
• W4384208072 hasAuthorship W4384208072A5046357357 @default.
• W4384208072 hasAuthorship W4384208072A5081589298 @default.
• W4384208072 hasAuthorship W4384208072A5090385327 @default.
• W4384208072 hasConcept C104317684 @default.
• W4384208072 hasConcept C11413529 @default.
• W4384208072 hasConcept C115961682 @default.
• W4384208072 hasConcept C138885662 @default.
• W4384208072 hasConcept C151989614 @default.
• W4384208072 hasConcept C153180895 @default.
• W4384208072 hasConcept C154945302 @default.
• W4384208072 hasConcept C163294075 @default.
• W4384208072 hasConcept C185592680 @default.
• W4384208072 hasConcept C2776182073 @default.
• W4384208072 hasConcept C2776401178 @default.
• W4384208072 hasConcept C28490314 @default.
• W4384208072 hasConcept C41008148 @default.
• W4384208072 hasConcept C41895202 @default.
• W4384208072 hasConcept C52622490 @default.
• W4384208072 hasConcept C55493867 @default.
• W4384208072 hasConcept C63479239 @default.
• W4384208072 hasConcept C88485024 @default.
• W4384208072 hasConcept C99498987 @default.
• W4384208072 hasConceptScore W4384208072C104317684 @default.
• W4384208072 hasConceptScore W4384208072C11413529 @default.
• W4384208072 hasConceptScore W4384208072C115961682 @default.
• W4384208072 hasConceptScore W4384208072C138885662 @default.
• W4384208072 hasConceptScore W4384208072C151989614 @default.
• W4384208072 hasConceptScore W4384208072C153180895 @default.
• W4384208072 hasConceptScore W4384208072C154945302 @default.
• W4384208072 hasConceptScore W4384208072C163294075 @default.
• W4384208072 hasConceptScore W4384208072C185592680 @default.
• W4384208072 hasConceptScore W4384208072C2776182073 @default.
• W4384208072 hasConceptScore W4384208072C2776401178 @default.
• W4384208072 hasConceptScore W4384208072C28490314 @default.
• W4384208072 hasConceptScore W4384208072C41008148 @default.
• W4384208072 hasConceptScore W4384208072C41895202 @default.
• W4384208072 hasConceptScore W4384208072C52622490 @default.
• W4384208072 hasConceptScore W4384208072C55493867 @default.
• W4384208072 hasConceptScore W4384208072C63479239 @default.
• W4384208072 hasConceptScore W4384208072C88485024 @default.
• W4384208072 hasConceptScore W4384208072C99498987 @default.
• W4384208072 hasFunder F4320321001 @default.
• W4384208072 hasFunder F4320333688 @default.
• W4384208072 hasLocation W43842080721 @default.
• W4384208072 hasOpenAccess W4384208072 @default.
• W4384208072 hasPrimaryLocation W43842080721 @default.
• W4384208072 hasRelatedWork W1520941986 @default.
• W4384208072 hasRelatedWork W1614028504 @default.
• W4384208072 hasRelatedWork W1847576989 @default.
• W4384208072 hasRelatedWork W2000534859 @default.
• W4384208072 hasRelatedWork W2027890689 @default.
• W4384208072 hasRelatedWork W2107577440 @default.
• W4384208072 hasRelatedWork W2164794243 @default.
• W4384208072 hasRelatedWork W2372023806 @default.
• W4384208072 hasRelatedWork W2381025269 @default.
• W4384208072 hasRelatedWork W2399955410 @default.
• W4384208072 hasVolume "141" @default.
• W4384208072 isParatext "false" @default.
• W4384208072 isRetracted "false" @default.
• W4384208072 workType "article" @default.