Academic Journal of Computing & Information Science, 2023, 6(5); doi: 10.25236/AJCIS.2023.060515.
Wuhan Qingchuan University, Wuhan, 430000, China
The extraction of speech features is the basis of speech signal processing, and the extraction of speech features is to obtain the parameters representing speech signals through the analysis of speech signals. The shape information of the signal can be extracted and processed by selecting appropriate structural elements and adopting mathematical morphological transformation. Selecting different structural elements will result in different shape transformation results, thus extracting the shape information of different components. In this paper, the feature extraction and analysis of speech signals are further studied based on FRFT (fractional fourier transform). The simulation results show that in different noise environments, such as when the speech is bathed with signal-to-noise ratio of 0dB and 10dB respectively, the recognition rate of this method is higher than that of the traditional parametric method. In this paper, based on the strength of FRFT signal components, the component signals are detected one by one in order, and then according to the detection results, the strongest component signals are removed from the observed signals in order to reduce their influence on the weak component signals, thus improving the effectiveness and reliability of multi-component signal detection and parameter estimation.
Fractional Fourier transform, Speech signal, Feature extraction
Yueying Zhai. Feature Extraction and Analysis of Speech Signal Based on Fractional Fourier Transform. Academic Journal of Computing & Information Science (2023), Vol. 6, Issue 5: 107-112. https://doi.org/10.25236/AJCIS.2023.060515.
 Dreuw P. Method for recognizing a voice context for a voice control function, method for ascertaining a voice control signal for a voice control function, and apparatus for executing the method. vol. 67, no. 26, pp. 34-58, 2018.
 Ikuma T, Mcwhorter A J, Adkins L, et al. Time-varying harmonic models for voice signal analysis. vol. 49, no. 23, pp. 44-67, 2022.
 Yu xing. Li, Xiao. Chen, Jing. Yu, Xiao hui. Yang, and Hui jun. Yang, The data-driven optimization method and its application in feature extraction of ship-radiated noise with sample entropy. Energies, vol. 12, no. 3, p. 359, 2019.
 Yu xing. Li, Long. Wang, Xue ping. Li, and Xiao hui. Yang, A novel linear spectrum frequency feature extraction technique for warship radio noise based on complete ensemble empirical mode decomposition with adaptive noise, duffing chaotic oscillator, and weighted-permutation entropy. Entropy, vol. 21, no. 5, p. 507, 2019.
 Zhi yan Han and Jian Wang, Dynamic feature extraction for speech signal based on formant curve and MUSIC. Chinese Control and Decision Conference (CCDC), Chongqing, China, pp. 403-407, 2017.
 Zhao R B, Rui Y B. Micro-Doppler feature extraction method under low signal-to-noise ratio. Information Technology, vol. 63, no. 10, pp. 15-65, 2017.
 Yan zhu Hu, Song Wang, Lei yuan Li. Feature Extraction Research in Signal Process with Phi-OTDR Technology[J]. Journal of Beijing University of Posts and Telecom, 2017, 40(6): 109-114.
 Hao T, Xue Y, Zheng Z. Feature Extraction of Noisy Signals Based on Short-Time Fractional Fourier Transform. Journal of Naval Aeronautical and Astronautical University, vol. 22, no. 13, pp. 25-29, 2019.
 Zhang Y, Zhang J, Tao R. Key Frame Extraction of Surveillance Video Based on Fractional Fourier Transform. Journal of Beijing University of Technology: English, vol. 30, no. 3, pp. 11-18, 2021.
 Zhang Y, Zhang J, Tao R. Key Frame Extraction of Surveillance Video Based on Fractional Fourier Transform. Journal of Beijing Institute of Technology, vol. 30, no. 3, pp. 311-321, 2021.
 Shao Y, Di L U, Yang G X. Application of Fractional Fourier Transform in Fault Diagnostics of Rolling Bearing. Journal of Harbin University of Science and Technology, vol. 11, no. 7, pp. 8-14, 2017.
 Wang S, Guo Y, Yang L. Research on sparsity of frequency modulated signal in fractional Fourier transform domain. Editorial Office of Opto-Electronic Engineering, vol. 52, no. 11, pp. 20-32, 2020.
 Daulappa Guranna BHALKE, Betsy RAJESH, Dattatraya Shankar BORMANE. Automatic Genre Classification Using Fractional Fourier Transform Based Mel Frequency Cepstral Coefficient and Timbral Features. Archives of acoustics: journal of Polish Academy of Sciences, vol. 42, no. 2, pp. 213-222, 2017.