FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2943754275> ?p ?o ?g. }

• W2943754275 abstract "With speech being a natural form of human communication in everyday life, speechprocessing technologies are gradually being incorporated into modern devices andmany applications such as Automatic Speech Recognition (ASR). Efficient ASR im-plementations are key for technologies which simplify service accessibility for clientssuch as automatic translation and dictation software. ASR task is known as aclient/server model. Various modes have been proposed for this approach includ-ing Network Speech Recognition (NSR) and Distributed Speech Recognition (DSR).In NSR, feature extraction and recognition task are performed at the server-side.The advantage of NSR is that no changes are required for the existing mobile tele-phony equipment and networks. Conversely, the ASR features are extracted fromthe speech signal at the client-side in DSR. Unlike the Linear Predictive Coding(LPC) parameters used in NSR, which are known to be more sensitive to the effectsof undesirable noise, DSR features can exploit human auditory effects which signi -cantly aid in speech recognition. The downside of DSR, however, is the requirementof a dedicated channel for the transmission process.This thesis presents novel signal processing algorithms that extract perceptually-motivated LPC parameters for computation at the client-side { which provide goodquality speech coding { and enable noise-robust ASR features at the server-side.These algorithms are demonstrated through three studies.The rst study introduces a proposed method for estimating perceptually-motivated LPC parameters { the Smoothed and Thresholded Power Spectrum Lin-ear Prediction (STPS-LP) analysis. The algorithm is based on the property of simultaneous masking found in the human auditory system to estimate noise-robust LPCparameters. The proposed method is evaluated and compared to three other LPanalysis methods: the conventional Autocorrelation Method (AM-LP), the Spec-tral Envelope Estimation Vocoder method (SEEVOC-LP), and the LP SpectrumModi cation method (LPSM-LP). Comparisons of the robustness and quality ofspeech demonstrated that the proposed STPS-LP method outperforms the threeother schemes.The second study investigates the quantisation performance of various LPC pa-rameters using different quantiser schemes. The proposed STPS-LP method pro-duced less quantisation distortion than the aforementioned methods.The third study builds on the proposed coefficients from the rst study, proposinga conversion algorithm for obtaining a set of noise-robust ASR features to be usedat the server-side. These cepstral-based features are called the STPS-LP CepstralCoefficients (STPS-LPCCs). The recognition performance using the STPS-LPCCsis improved in comparison to that using the AM-LP, SEEVOC-LP, and LPSM-LPcepstral coefficients, under clean and noisy conditions in the baseline, matched, andmismatched models.This research proposal provides two main advantages over the conventional NSRand DSR schemes: (i) the perceptually-motivated LPC parameters are used forspeech coding, and (ii) no dedicated communications channel is required since theexisting LPC bitstream in speech coders is used for transmitting the features. Thesebene ts make the proposed method greatly applicable to current mobile telephonynetworks and should improve the user experience when interacting with ASR ser-vices." @default.
• W2943754275 created "2019-05-09" @default.
• W2943754275 creator A5051298320 @default.
• W2943754275 date "2018-03-01" @default.
• W2943754275 modified "2023-09-27" @default.
• W2943754275 title "Perceptually-Motivated Speech Parameters for Efficient Coding and Noise-Robust Cepstral-Based ASR Features" @default.
• W2943754275 doi "https://doi.org/10.25904/1912/1159" @default.
• W2943754275 hasPublicationYear "2018" @default.
• W2943754275 type Work @default.
• W2943754275 sameAs 2943754275 @default.
• W2943754275 citedByCount "0" @default.
• W2943754275 crossrefType "dissertation" @default.
• W2943754275 hasAuthorship W2943754275A5051298320 @default.
• W2943754275 hasConcept C13895895 @default.
• W2943754275 hasConcept C154945302 @default.
• W2943754275 hasConcept C163294075 @default.
• W2943754275 hasConcept C204201278 @default.
• W2943754275 hasConcept C2776182073 @default.
• W2943754275 hasConcept C28490314 @default.
• W2943754275 hasConcept C41008148 @default.
• W2943754275 hasConcept C59883199 @default.
• W2943754275 hasConcept C61328038 @default.
• W2943754275 hasConceptScore W2943754275C13895895 @default.
• W2943754275 hasConceptScore W2943754275C154945302 @default.
• W2943754275 hasConceptScore W2943754275C163294075 @default.
• W2943754275 hasConceptScore W2943754275C204201278 @default.
• W2943754275 hasConceptScore W2943754275C2776182073 @default.
• W2943754275 hasConceptScore W2943754275C28490314 @default.
• W2943754275 hasConceptScore W2943754275C41008148 @default.
• W2943754275 hasConceptScore W2943754275C59883199 @default.
• W2943754275 hasConceptScore W2943754275C61328038 @default.
• W2943754275 hasLocation W29437542751 @default.
• W2943754275 hasOpenAccess W2943754275 @default.
• W2943754275 hasPrimaryLocation W29437542751 @default.
• W2943754275 hasRelatedWork W101328856 @default.
• W2943754275 hasRelatedWork W1609439988 @default.
• W2943754275 hasRelatedWork W1990155902 @default.
• W2943754275 hasRelatedWork W1995828869 @default.
• W2943754275 hasRelatedWork W2057826000 @default.
• W2943754275 hasRelatedWork W2096638934 @default.
• W2943754275 hasRelatedWork W2100692852 @default.
• W2943754275 hasRelatedWork W2105839915 @default.
• W2943754275 hasRelatedWork W2117441840 @default.
• W2943754275 hasRelatedWork W2132097179 @default.
• W2943754275 hasRelatedWork W2140268222 @default.
• W2943754275 hasRelatedWork W2149119909 @default.
• W2943754275 hasRelatedWork W2153569794 @default.
• W2943754275 hasRelatedWork W2233657726 @default.
• W2943754275 hasRelatedWork W2556506241 @default.
• W2943754275 hasRelatedWork W2579473097 @default.
• W2943754275 hasRelatedWork W269674882 @default.
• W2943754275 hasRelatedWork W2920974976 @default.
• W2943754275 hasRelatedWork W2965809951 @default.
• W2943754275 hasRelatedWork W2183527285 @default.
• W2943754275 isParatext "false" @default.
• W2943754275 isRetracted "false" @default.
• W2943754275 magId "2943754275" @default.
• W2943754275 workType "dissertation" @default.