MFCC based real-time speech reproduction and recognition using distributed acoustic sensing technology

Abstract

The distributed acoustic sensing technology was used for real-time speech reproduction and recognition, in which the voiceprint can be extracted by the Mel frequency cepstral coefficient (MFCC) method. A classic ancient Chinese poem “You Zi Yin”, also called “A Traveler’s Song”, was analyzed both in time and frequency domains, where its real-time reproduction was achieved with a 116.91 ms time delay. The smaller scaled MFCC₀ at 1/12 of MFCC matrix was taken as a feature vector of each line against the ambient noise, which provides a recognition method via cross-correlation among the 6 original and recovered verse pairs. The averaged cross-correlation coefficient of the matching pairs is calculated to be 0.580 6 higher than 0.188 3 of the nonmatched pairs, promising an accurate and fast method for real-time speech reproduction and recognition over a passive optical fiber.

References

[1] WittjeR. The electrical imagination: sound analogies, equivalent circuits, and the rise of electroacoustics, 1863–1939[J]. Osiris, 2013, 28(1):40-63

[2] HuangY, ZhouX. Non-reciprocal sound transmission in electro-acoustic systems with time-modulated circuits[J]. Acta mechanica solida sinica, 2022, 35(6):940-948

[3] SherifM M, KhakimovaE M, TanksJ, et al.. Cyclic flexural behavior of hybrid SMA/steel fiber reinforced concrete analyzed by optical and acoustic techniques[J]. Composite structures, 2018, 201: 248-260

[4] JoeH E, YunH, JoS H, et al.. A review on optical fiber sensors for environmental monitoring[J]. International journal of precision engineering and manufacturing-green technology, 2018, 5(1):173-191

[5] HubbardP G, XuJ, ZhangS, et al.. Dynamic structural health monitoring of a model wind turbine tower using distributed acoustic sensing (DAS)[J]. Journal of civil structural health monitoring, 2021, 11(3):833-849

[6] FERNANDZE-RUIZ M R, SOTO M A, WILLIAMS E F, et al. Distributed acoustic sensing for seismic activity monitoring[J]. APL photonics, 2020, 5(3).

[7] MocciaM, PiscoM, CutoloA, et al.. Op-to-acoustic behavior of coated fiber Bragg gratings[J]. Optics express, 2011, 19(20):18842-18860

[8] Macia-SanahujaC, LamelaH, Garcia-SoutoJ A. Fiber optic interferometric sensor for acoustic detection of partial discharges[J]. Journal of optical technology, 2007, 74(2):122-126

[9] XiongJ, WangZ, JiangJ, et al.. High sensitivity and large measurable range distributed acoustic sensing with Rayleigh-enhanced fiber[J]. Optics letters, 2021, 46(11):2569-2572

[10] FangG, XuT, FenfS, et al.. Phase-sensitive optical time domain reflectometer based on phase-generated carrier algorithm[J]. Journal of lightwave technology, 2015, 33(13):2811-2816

[11] WangZ, ZhangL, WangS, et al.. Coherent Φ-OTDR based on I/Q demodulation and homodyne detection[J]. Optics express, 2016, 24(2):853-858

[12] TuG, ZhangX, ZhangY, et al.. The development of an Φ-OTDR system for quantitative vibration measurement[J]. IEEE photonics technology letters, 2015, 27(12):1349-1352

[13] WangS, JiangJ, WangS, et al.. GPU-based fast processing for a distributed acoustic sensor using an LFM pulse[J]. Applied optics, 2020, 59(35):11098-11103

[14] ZhuK, ZhouB, WuH, et al.. Multipath distributed acoustic sensing system based on phase-sensitive optical time-domain reflectometry with frequency division multiplexing technique[J]. Optics and lasers in engineering, 2021, 142: 106593

[15] ZhangX, QiaoW, SunZ, et al.. A distributed optical fiber sensing system for synchronous vibration and loss measurement[J]. Optoelectronics letters, 2016, 12(5):375-378

[16] LuY, ZhuT, ChenL, et al.. Distributed vibration sensor based on coherent detection of phase-OTDR[J]. Journal of lightwave technology, 2010, 28(22):3243-3249

[17] DongY, ChenX, LiuE, et al.. Quantitative measurement of dynamic nanostrain based on a phase-sensitive optical time domain reflectometer[J]. Applied optics, 2016, 55(28):7810-7815

[18] MasoudiA, BelalM, NewsonT P. Distributed optical fiber audible frequency sensor[C]. 23rd International Conference on Optical Fiber Sensors, June 2–6, 2014, Santander, Spain, 2014, Washington, SPIE: 537-540

[19] FranciscangelisC, MargulisW, KjellbergL, et al.. Real-time distributed fiber microphone based on phase-OTDR[J]. Optics express, 2016, 24(26):29597-29602

[20] WuY, GanJ, LiQ, et al.. Distributed fiber voice sensor based on phase-sensitive optical time-domain reflectometry[J]. IEEE photonics journal, 2015, 7(6):1-10

[21] ZhangP, VenketeswaranA, WrightR, et al.. Feature extraction for pipeline defects inspection based upon distributed acoustic fiber optic sensing data[C]. Fiber Optic Sensors and Applications XVIII, April 3–June 12, 2022, Virtual, 2022, Washington, SPIE: 14-29

[22] TabjulaJ, SharmaJ. Feature extraction techniques for noisy distributed acoustic sensor data acquired in a wellbore[J]. Applied optics, 2023, 62(16):E51-E61

[23] NingF, ChengZ, MengD, et al.. A framework combining acoustic features extraction method and random forest algorithm for gas pipeline leak detection and classification[J]. Applied acoustics, 2021, 182: 108255

[24] WuH, WangC, LiuX, et al.. Intelligent target recognition for distributed acoustic sensors by using both manual and deep features[J]. Applied optics, 2021, 60(23):6878-6887

[25] JiangF, LiH, ZhangZ, et al.. An event recognition method for fiber distributed acoustic sensing systems based on the combination of MFCC and CNN[C]. 2017 International Conference on Optical Instruments and Technology: Advanced Optical Sensors and Applications, October 28–30, 2017, Beijing, China, 2018, Washington, SPIE: 15-21

[26] ShiY, LiuX, WeiC. An event recognition method based on MFCC, superposition algorithm and deep learning for buried distributed optical fiber sensors[J]. Optics communications, 2022, 522: 128647

[27] ShangY, YangJ, ChenW, et al.. Speech signal enhancement based on deep learning in distributed acoustic sensing[J]. Optics express, 2023, 31(3):4067-4079

[28] BencharifB A E, BayarS, ÖzkanE. Parallel implementation of distributed acoustic sensor acquired signals: detection, processing, and classification[J]. Journal of applied remote sensing, 2022, 16(2):024504-024504

[29] AYVAZ U, GURULER H, KHAN F, et al. Automatic speaker recognition using mel-frequency cepstral coefficients through machine learning[J]. CMC-computers materials & continua, 2022, 71(3).

[30] ArpithaY, MashumathiG L, BalajiN. Spectrogram analysis of ECG signal and classification efficiency using MFCC feature extraction technique[J]. Journal of ambient intelligence and humanized computing, 2022, 13(2):757-767

[31] GANCHEV T, FAKOTAKIS N, KOKKINAKIS G. Comparative evaluation of various MFCC implementations on the speaker verification task[C]//Proceedings of the SPECOM, October 17–19, 2005, Patras, Greece. Moscow, 2005: 191–194.

[32] BlotekjaerK. Fundamental noise sources that limit the ultimate resolution of fiber optic sensors[C]. Optical and Fiber Optic Sensor Systems, September 16–19, 1998, Beijing, China, 1998, Washington, SPIE: 1-12