Matches in SemOpenAlex for { <https://semopenalex.org/work/W4380987202> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W4380987202 endingPage "61957" @default.
- W4380987202 startingPage "61950" @default.
- W4380987202 abstract "Speech emotion recognition (SER) has become a major area of investigation in human-computer interaction. Conventionally, SER is formulated as a classification problem that follows a common methodology: (i) extracting features from speech signals; and (ii) constructing an emotion classifier using extracted features. With the advent of deep learning, however, the former stage is integrated into the latter. That is to say, deep neural networks (DNNs), which are trained using log-Mel spectrograms (LMS) of audio waveforms, extract discriminative features from LMS. A critical issue, and one that is often overlooked, is that this procedure is done without relating the choice of LMS parameters to the performance of the trained DNN classifiers. It is commonplace in SER studies that practitioners assume some “usual” values for these parameters and devote major efforts to training and comparing various DNN architectures. In contrast with this common approach, in this work we choose a single lightweight pre-trained architecture, namely, SqueezeNet, and shift our main effort into tuning LMS parameters. Our empirical results using three publicly available SER datasets show that: (i) parameters of LMS can considerably affect the performance of DNNs; and (ii) by tuning LMS parameters, highly competitive classification performance can be achieved. In particular, treating LMS parameters as hyperparameters and tuning them led to ~23%, ~10%, and ~11% improvement in contrast with the use of “usual” values of LMS parameters in EmoDB, IEMOCAP, and SAVEE datasets, respectively." @default.
- W4380987202 created "2023-06-17" @default.
- W4380987202 creator A5040086890 @default.
- W4380987202 creator A5057434469 @default.
- W4380987202 creator A5092188094 @default.
- W4380987202 date "2023-01-01" @default.
- W4380987202 modified "2023-10-14" @default.
- W4380987202 title "On the Effect of Log-Mel Spectrogram Parameter Tuning for Deep Learning-Based Speech Emotion Recognition" @default.
- W4380987202 cites W175750906 @default.
- W4380987202 cites W2097117768 @default.
- W4380987202 cites W2146334809 @default.
- W4380987202 cites W2166725142 @default.
- W4380987202 cites W2194775991 @default.
- W4380987202 cites W2485020921 @default.
- W4380987202 cites W2777186991 @default.
- W4380987202 cites W2799331981 @default.
- W4380987202 cites W2884739346 @default.
- W4380987202 cites W2885005742 @default.
- W4380987202 cites W2889191349 @default.
- W4380987202 cites W2959133507 @default.
- W4380987202 cites W2960833983 @default.
- W4380987202 cites W2970737019 @default.
- W4380987202 cites W2972498864 @default.
- W4380987202 cites W3015240477 @default.
- W4380987202 cites W3020839184 @default.
- W4380987202 cites W3040201034 @default.
- W4380987202 cites W3112103486 @default.
- W4380987202 cites W3132191748 @default.
- W4380987202 cites W3136524425 @default.
- W4380987202 cites W3164680725 @default.
- W4380987202 cites W3186192207 @default.
- W4380987202 cites W3212554192 @default.
- W4380987202 cites W4226153164 @default.
- W4380987202 cites W4285106979 @default.
- W4380987202 cites W4288765755 @default.
- W4380987202 cites W3192216513 @default.
- W4380987202 doi "https://doi.org/10.1109/access.2023.3287093" @default.
- W4380987202 hasPublicationYear "2023" @default.
- W4380987202 type Work @default.
- W4380987202 citedByCount "0" @default.
- W4380987202 crossrefType "journal-article" @default.
- W4380987202 hasAuthorship W4380987202A5040086890 @default.
- W4380987202 hasAuthorship W4380987202A5057434469 @default.
- W4380987202 hasAuthorship W4380987202A5092188094 @default.
- W4380987202 hasBestOaLocation W43809872021 @default.
- W4380987202 hasConcept C108583219 @default.
- W4380987202 hasConcept C119857082 @default.
- W4380987202 hasConcept C153180895 @default.
- W4380987202 hasConcept C154945302 @default.
- W4380987202 hasConcept C2776502983 @default.
- W4380987202 hasConcept C28490314 @default.
- W4380987202 hasConcept C2984842247 @default.
- W4380987202 hasConcept C41008148 @default.
- W4380987202 hasConcept C45273575 @default.
- W4380987202 hasConcept C50644808 @default.
- W4380987202 hasConcept C8642999 @default.
- W4380987202 hasConcept C95623464 @default.
- W4380987202 hasConcept C97931131 @default.
- W4380987202 hasConceptScore W4380987202C108583219 @default.
- W4380987202 hasConceptScore W4380987202C119857082 @default.
- W4380987202 hasConceptScore W4380987202C153180895 @default.
- W4380987202 hasConceptScore W4380987202C154945302 @default.
- W4380987202 hasConceptScore W4380987202C2776502983 @default.
- W4380987202 hasConceptScore W4380987202C28490314 @default.
- W4380987202 hasConceptScore W4380987202C2984842247 @default.
- W4380987202 hasConceptScore W4380987202C41008148 @default.
- W4380987202 hasConceptScore W4380987202C45273575 @default.
- W4380987202 hasConceptScore W4380987202C50644808 @default.
- W4380987202 hasConceptScore W4380987202C8642999 @default.
- W4380987202 hasConceptScore W4380987202C95623464 @default.
- W4380987202 hasConceptScore W4380987202C97931131 @default.
- W4380987202 hasFunder F4320327695 @default.
- W4380987202 hasLocation W43809872021 @default.
- W4380987202 hasOpenAccess W4380987202 @default.
- W4380987202 hasPrimaryLocation W43809872021 @default.
- W4380987202 hasRelatedWork W2112343299 @default.
- W4380987202 hasRelatedWork W2353457699 @default.
- W4380987202 hasRelatedWork W2791691546 @default.
- W4380987202 hasRelatedWork W2905846897 @default.
- W4380987202 hasRelatedWork W2950066684 @default.
- W4380987202 hasRelatedWork W3047644063 @default.
- W4380987202 hasRelatedWork W3158264953 @default.
- W4380987202 hasRelatedWork W4283697347 @default.
- W4380987202 hasRelatedWork W4298388782 @default.
- W4380987202 hasRelatedWork W4307195028 @default.
- W4380987202 hasVolume "11" @default.
- W4380987202 isParatext "false" @default.
- W4380987202 isRetracted "false" @default.
- W4380987202 workType "article" @default.