An Optimal Feature Parameter Set Based on Gated Recurrent Unit Recurrent Neural Networks for Speech Segment Detection
Open accessThe DOI matches its Crossref record; 43 of the 43 references it cites were found in a registry or an index.
Abstract
Speech segment detection based on gated recurrent unit (GRU) recurrent neural networks for the Kurdish language was investigated in the present study. The novelties of the current research are the utilization of a GRU in Kurdish speech segment detection, creation of a unique database from the Kurdish language, and optimization of processing parameters for Kurdish speech segmentation. This study is the first attempt to find the optimal feature parameters of the model and to form a large Kurdish vocabulary dataset for a speech segment detection based on consonant, vowel, and silence (C/V/S) discrimination. For this purpose, four window sizes and three window types with three hybrid feature vector techniques were used to describe the phoneme boundaries. Identification of the phoneme boundaries using a GRU recurrent neural network was performed with six different classification algorithms for the C/V/S discrimination. We have demonstrated that the GRU model has achieved outstanding speech segmentation performance for characterizing Kurdish acoustic signals. The experimental findings of the present study show the significance of the segment detection of speech signals by effectively utilizing hybrid features, window sizes, window types, and classification models for Kurdish speech.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“RNN is mainly applied in the area of speech processing and NLP contexts [66, 67].”
From Review of deep learning: concepts, CNN architectures, challenges, applications, future directions (2021) · cites this paper as [66]
Citations
Cites
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318 · cited 1×
“Alternatively, LSTM is a slightly more complex structure variation of the GRU [14].”
Data Mining: Practical Machine Learning Tools and Techniques
(2011) · Elsevier eBooks · doi:10.1016/c2009-0-19715-5 · Cited by 25,437
Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
Chung, J., Gulcehre, C., Cho, K., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1412.3555 · Cited by 11,008 · cited 1×
“Therefore, it should exhibit a better performance in comparison with conventional machine learning techniques [13].”
On the use of windows for harmonic analysis with the discrete Fourier transform
Harris, F. J. (1978) · Proceedings of the IEEE · doi:10.1109/proc.1978.10837 · Cited by 7,326 · cited 1×
“The most commonly used windowing functions are the Hamming, Hanning, and rectangular windowing techniques [44,45].”
Framewise phoneme classification with bidirectional LSTM and other neural network architectures
Graves, A., & Schmidhuber, J. (2005) · Neural Networks · doi:10.1016/j.neunet.2005.06.042 · Cited by 5,839
Praat, a system for doing phonetics by computer
Boersma, P. (2002) · UvA-DARE (University of Amsterdam) · Cited by 4,417
Speech emotion recognition using deep 1D & 2D CNN LSTM networks
Zhao, J., Mao, X., & Chen, L. (2019) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2018.08.035 · Cited by 1,044
New insights into the noise reduction Wiener filter
Chen, J., Benesty, J., Huang, Y., & Doclo, S. (2006) · IEEE Transactions on Audio, Speech and Language Processing · doi:10.1109/tsa.2005.860851 · Cited by 706
Performance Evaluation of Deep Neural Networks Applied to Speech Recognition: RNN, LSTM and GRU
Shewalkar, A., Nyavanandi, D., & Ludwig, S. A. (2019) · Journal of Artificial Intelligence and Soft Computing Research · doi:10.2478/jaiscr-2019-0006 · Cited by 576 · cited 1×
“The long short ‐ term memory (LSTM) and GRU were assessed for the purpose of comparing the performances they exhibited upon a reduced Technology Entertainment Design ‐ Laboratoire Informatique de l’Université du Maine (TED ‐ LIUM) speech data set [6].”
Light Gated Recurrent Units for Speech Recognition
Ravanelli, M., Brakel, P., Omologo, M., & Bengio, Y. (2018) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2017.2762739 · Cited by 411 · cited 1×
“In Ravanelli et al. [5], GRUs were reviewed and a compact single ‐ gate model replacing a hyperbolic tangent with rectified linear unit activations having a simplified architecture was suggested for the purpose of automatic speech recognition (ASR).”
A novel approach for MFCC feature extraction
Hossan, M. A., Memon, S., & Gregory, M. A. (2010) · 2010 4th International Conference on Signal Processing and Communication Systems · doi:10.1109/icspcs.2010.5709752 · Cited by 300 · cited 1×
“Delta‐MFCC (D‐MFCC) and the delta‐delta‐MFCC (DD‐MFCC) are the first‐order and the second‐order derivatives of the MFCC, respectively [50].”
Convolutional RNN: An enhanced model for extracting features from sequential data
Keren, G., & Schuller, B. (2016) · 2016 International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.2016.7727636 · Cited by 155 · cited 1×
“These models were applied to different speech processing tasks to improve the accuracy [25–29].”
Hierarchical classification of audio data for archiving and retrieving
Zhang, T., & Kuo, C. ‐. J. (1999) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.1999.757472 · Cited by 131 · cited 2×
“It is also a measure of the number of times in a particular time interval/frame when the amplitude of speech signals passes through a zero value [46–48].”
WekaDeeplearning4j: A deep learning package for Weka based on Deeplearning4j
Lang, S., Bravo-Márquez, F., Beckham, C., Hall, M., & Frank, E. (2019) · Knowledge-Based Systems · doi:10.1016/j.knosys.2019.04.013 · Cited by 131 · cited 1×
“For this operation, the PRAAT 6.0.49 tool [31], MATLAB R2018a (MathWorks, Natick, MA 01760 ‐ 2098, USA), Keras Library [32], Weka 3.9.3 [33], and WekaDeeplearning4j [34] are used.”
InfluentialA system for acoustic-phonetic analysis of continuous speech
Weinstein, C. J., McCandless, S., Mondshein, L., & Zue, V. W. (1975) · IEEE Transactions on Acoustics Speech and Signal Processing · doi:10.1109/tassp.1975.1162651 · Cited by 90 · cited 1×
“In Weinstein et al. [17], the development of a system to conduct the acoustic–phonetic analysis of continuous speech was performed with the aim of serving as a part of an automatic speech understanding system.”
Automatic Speech Segmentation for Concatenative Inventory Selection
Ljolje, A., Hirschberg, J., & van Santen, J. P. H. (1997) · Springer eBooks · doi:10.1007/978-1-4612-1894-4_24 · Cited by 57
Acoustic-phonetic features for the automatic classification of stop consonants
Ali, A. M. A., der Spiegel, J. V., & Müeller, P. (2001) · IEEE Transactions on Speech and Audio Processing · doi:10.1109/89.966086 · Cited by 53 · cited 1×
“In Ali et al. [19], the acoustic and phonetic properties of the American English stop consonants were examined.”
An Ensemble Model for Multi-Level Speech Emotion Recognition
Zheng, C., Wang, C., & Jia, N. (2019) · Applied Sciences · doi:10.3390/app10010205 · Cited by 52 · cited 1×
“In Zheng et al. [8], general emotion features were produced in speech signals from various angles, and an ensemble learning model was employed for the purpose of carrying out emotion recognition tasks.”
A preliminary statistical evaluation of manual and automatic segmentation discrepancies
Cosi, P., Falavigna, D., & Omologo, M. (1991) · European Conference on Speech Communication and Technology (Eurospeech) · doi:10.21437/eurospeech.1991-183 · Cited by 44 · cited 1×
“The manual segmentation method is considered to exhibit more accurate performance in comparison with automatic segmentation [39–41].”
Vowels, consonants, speech, and nonspeech.
Ades, A. E. (1977) · Psychological Review · doi:10.1037/0033-295x.84.6.524 · Cited by 39
End-to-End Mandarin Speech Recognition Combining CNN and BLSTM
Wang, D., Wang, X., & Lv, S. (2019) · Symmetry · doi:10.3390/sym11050644 · Cited by 37
Optimization of processing parameters for the analysis and detection of embolic signals
Aydın, N., & Markus, H. S. (2000) · European Journal of Ultrasound · doi:10.1016/s0929-8266(00)00104-x · Cited by 36 · cited 1×
“The purpose of this step was to eliminate discontinuities at the beginning and end of each frame [43].”
Wang, Y.-H., Chung, C.-T., & Lee, H.-Y. (2017) · Interspeech 2017 · doi:10.21437/interspeech.2017-877 · Cited by 36 · cited 2×
“In Wang et al. [12], it is reported that GRU forget gate activations in trained recurrent acoustic neural networks correlate very well with phoneme boundaries in speech activation, which ensures the validation of numerous approaches to speech recognition and other sequence modeling problems,…”
Phonetic classification using multi-layer perceptrons
Leung, H. C., & Zue, V. W. (2002) · International Conference on Acoustics, Speech, and Signal Processing · doi:10.1109/icassp.1990.115765 · Cited by 35 · cited 1×
“[18], a segmental approach was employed for the purpose of phonetic recognition.”
Phoneme Boundary Detection using Deep Bidirectional LSTMs
Franke, J., Müller, M., Hamaloui, F., Stüker, S., & Waibel, A. (2016) · KITopen · doi:10.5445/ir/1000166279 · Cited by 29
A NOVEL METHOD FOR ARABIC CONSONANT/VOWEL SEGMENTATION USING WAVELET TRANSFORM
Nazmy, T. M., Gadallah, M., & Abdelhamid, A. A. (2005) · Egyptian Computer Science Journal · Cited by 22
Techniques for accurate automatic annotation of speech waveforms
Cox, S., Brady, R., & Jackson, P. E. (1998) · International Conference on Spoken Language Processing (ICSLP) · doi:10.21437/icslp.1998-22 · Cited by 19
Convolutional RNN: an Enhanced Model for Extracting Features from Sequential DataPreprint
Keren, G., & Schuller, B. W. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1602.05875 · Cited by 18 · cited 1×
“These models were applied to different speech processing tasks to improve the accuracy [25–29].”
Perspectives and applications of machine learning for evolutionary developmental biology
Feltes, B. C., Grisci, B. I., Poloni, J. D. F., & Dorn, M. (2018) · Molecular Omics · doi:10.1039/c8mo00111a · Cited by 17 · cited 1×
“The above ‐ mentioned operations are performed repeatedly over tens or hundreds of layers, with every layer learning for the purpose of detecting various features in the input data [53].”
Wang, Y., Chung, C.-T., & Lee, H.-Y. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1703.07588 · Cited by 16 · cited 2×
“In Wang et al. [12], it is reported that GRU forget gate activations in trained recurrent acoustic neural networks correlate very well with phoneme boundaries in speech activation, which ensures the validation of numerous approaches to speech recognition and other sequence modeling problems,…”
Phoneme segmentation of continuous speech using multi-layer perceptron
Suh, Y., & Lee, Y. (1996) · International Conference on Spoken Language Processing (ICSLP) · doi:10.21437/icslp.1996-332 · Cited by 12
Unconstrained Speech Segmentation using Deep Neural Networks
van Vuuren, V. Z., ten Bosch, L., & Niesler, T. (2015) · International Conference on Pattern Recognition Applications and Methods · doi:10.5220/0005201802480254 · Cited by 11
Segmentation of Continuous Speech into Consonant and Vowel Units using Formant Frequencies
AnanthaNatarajan, V., & Jothilakshmi, S. (2012) · International Journal of Computer Applications · doi:10.5120/8968-3185 · Cited by 11 · cited 1×
“In Natarajan and Jothilakshmi [20], the segmentation of the continuous speech into smaller speech units was performed, and every unit was classified as a consonant or vowel by utilizing the formant frequencies and support vector machines (SVMs).”
Performance Analysis of Windowing Techniques in Automatic Speech Signal Segmentation
Chithra, P. L., & Aparna, R. (2015) · Indian Journal of Science and Technology · doi:10.17485/ijst/2015/v8i29/83616 · Cited by 11 · cited 1×
“The most commonly used windowing functions are the Hamming, Hanning, and rectangular windowing techniques [44,45].”
The extraction and simulation of Mel frequency cepstrum speech parameters
Xu, H., Zhang, X., & Jia, L. (2012) · International Conference on Systems and Informatics (ICSAI) · doi:10.1109/icsai.2012.6223385 · Cited by 10 · cited 1×
“(MFCC) is known to mimic human hearing system [30].”
Detection times for vowels versus consonants
van Ooyen, B., Cutler, A., & Norris, D. G. (1991) · European Conference on Speech Communication and Technology (Eurospeech) · doi:10.21437/eurospeech.1991-270 · Cited by 9 · cited 1×
“[22], two tests with vowels and consonants as the targets of phoneme detection in real words were performed.”
Design and implementation of voice command using MFCC and HMMs method
Sidiq, M., Wirayuda, T. A. B., & Saadah, S. (2015) · doi:10.1109/icoict.2015.7231454 · Cited by 6
Nasal Speech Sounds Detection Using Connectionist Temporal Classification
Cerňak, M., & Tong, S. (2018) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2018.8462149 · Cited by 5 · cited 2×
“In Cernak and Tong [7], a solution was suggested with the aim of training a phone attribute detector without phone alignment by utilizing end ‐ to ‐ end phone attribute modeling on the basis of the connectionist temporal classification.”
A Practical Singing Voice Detection System Based on GRU-RNN
Chen, Z., Zhang, X., Deng, J. Y., Li, J., Jiang, Y. L., & Li, W. (2019) · Lecture notes in electrical engineering · doi:10.1007/978-981-13-8707-4_2 · Cited by 4 · cited 1×
“[9], a practical approach with three steps was introduced for singing voice detection on the basis of a GRU.”
Study on Feature Subspace of Archetypal Emotions for Speech Emotion RecognitionPreprint
Ma, X., Wu, Z., Jia, J., Xu, M., Meng, H. M. L., & Cai, L. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1611.05675 · Cited by 3 · cited 1×
“These models were applied to different speech processing tasks to improve the accuracy [25–29].”
Phoneme segmentation using deep learning for speech synthesis
Lee, Y. H., Jong-Yeol, Y., Cho, C., & Jung, H. (2018) · ACM Conference on Research in Adaptive and Convergent Systems (RACS) · doi:10.1145/3264746.3264801 · Cited by 3 · cited 1×
“[15] suggested phoneme segmentation by utilizing cross‐entropy loss with connectionist temporal classification loss in deep speech architecture for the purpose of performing speech synthesis.”
Measurement of Formant Frequency for Consonant-Vowel type Bodo words for acustic analysis
Sharma, U. (2014) · doi:10.1109/icdmic.2014.6954251 · Cited by 2 · cited 1×
“The placement of vowels and consonants in a word of any language is very significant for identifying various types of words [3].”
Remarks on Vowels and Consonants in Kurmanji
Gündoğdu, S. (2016) · Anemon Muş Alparslan Üniversitesi Sosyal Bilimler Dergisi · doi:10.18506/anemon.16606 · Cited by 2 · cited 1×
“The phonemes of Kurdish are presented in Table 1, in which Kurdish letters, International Phonetic Alphabet (IPA) notations, and phonetic descriptions are given [35–37].”
Research works printed but not linked: 9
- Sakran, 2017, A review: Automatic Speech Segmentation, IJCSMC, 6, 308
- Artuner, H. (1994). The Design and Implementation of a Turkish Speech Phoneme Clustering System. [Ph.D. Thesis, Ankara University].
- Goodfellow, I., Bengio, Y., and Courville, A. (2016). Deep Learning, MIT Press. Available online: http://www.deeplearningbook.org.
- Li, C., Ma, X., Jiang, B., Li, X., Zhang, X., Liu, X., Cao, Y., Kannan, A., and Zhu, Z. (2017). Deepspeaker: An end-to-end neural speaker embedding system. arXiv.
- Jain, N., and Kaushik, D. (2014, January 15). Gender voice recognition through speech analysis with higher accuracy. Proceedings of the 8th International Conference on Advance Computing and Communication Technology, Panipat, Haryana, India.
- Hemakumar, 2014, Automatic segmentation of Kannada speech signal into syllable and sub-words: Noised and noiseless signals, Int. J. Sci. Eng. Res., 5, 1707
- Kalamani, 2015, Hybrid speech segmentation algorithm for continuous speech recognition, Int. J. Appl. Inf. Commun. Eng., 1, 39
- Rana, R. (2016). Gated recurrent unit (GRU) for emotion classification from noisy speech. arXiv.
- Misra, 2014, Review of System Identification Using Neural Network Techniques, Int. J. Electr. Electron. Data Commun., 2, 13
Cited by
Review of deep learning: concepts, CNN architectures, challenges, applications, future directions
Alzubaidi, L., Zhang, J., Humaidi, A. J., Al-Dujaili, A., Duan, Y., Al-Shamma, O., Santamaría, J., Fadhel, M. A., Al-Amidie, M., & Farhan, L. (2021) · Journal of Big Data · doi:10.1186/s40537-021-00444-8 · Cited by 7,906
Artificial intelligence-based traffic flow prediction: a comprehensive review
Sayed, S. A., Abdel-Hamid, Y., & Hefny, H. A. (2023) · Journal of Electrical Systems and Information Technology · doi:10.1186/s43067-023-00081-6 · Cited by 158
A improved pooling method for convolutional neural networks
Zhao, L., & Zhang, Z. (2024) · Scientific Reports · doi:10.1038/s41598-024-51258-6 · Cited by 239
A novel hybrid deep neural network model for short‐term electricity price forecasting
Huang, C., Shen, Y., Chen, Y., & Chen, H. (2020) · International Journal of Energy Research · doi:10.1002/er.5945 · Cited by 133
Multiple-Input Deep Convolutional Neural Network Model for COVID-19 Forecasting in ChinaPreprint
Huang, C., Chen, Y., Ma, Y., & Kuo, P. (2020) · medRxiv · doi:10.1101/2020.03.23.20041608 · Cited by 124
Metadata
- Authors
- BATUR DİNLER, Ö., & AYDIN, N.
- Affiliations
- Computer Engineering, Siirt University, Siirt 56100, Turkey
- Computer Engineering, Yildiz Technical University, Istanbul 34220, Turkey
- ORCID
- AYDIN, Nizamettin
- Venue
- Applied Sciences
- Publisher
- MDPI AG
- Published
- 2020-02-13
- Volume
- 10
- Issue
- 4
- Pages
- 1273
- ISSN
- 2076-3417 (ISSN-L)
- OpenAlex
- W3006343578
- Cited by
- 45
- References
- 41
- Citations by year
- 2026: 3 · 2025: 3 · 2024: 8 · 2023: 8 · 2022: 11 · 2021: 6 · 2020: 6
- FWCI
- 3.04
- Type
- journal article
- Language
- en
- Open access
- gold
- dx.doi.org
- Licence
- Locations
- Applied Sciences (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Physical Sciences › Computer Science › Signal Processing › Speech and Audio Processing
- Physical Sciences › Computer Science › Signal Processing › Music and Audio Processing
- Keywords
- Kurdish language, gated recurrent unit, phoneme boundary detection, window size, hybrid feature vector, speech segmentation
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.