Takahiro Fukumori

Dr. Takahiro Fukumori


Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
30 April 2018 in IEEE/ACM Transactions on Audio, Speech, and Language Processing
Verified
Journal Article
Closed access
YWYukoh Wakabayashi
Yukoh Wakabayashi
TFTakahiro Fukumori
Takahiro Fukumori
Masato NakayamaMasato Nakayama
Masato Nakayama
TNTakanobu Nishiura
Takanobu Nishiura
YYYoichi Yamashita
Yoichi Yamashita
Speech enhancement has been widely investigated for several decades, but by modifying only the amplitude spectrum of a speech signal, ignoring the phase spectrum, which has been regarded as an unimportant feature. However, it was recently reported that the phase spectrum plays an important role in speech quality and intelligibility. In this paper, we propose a phase reconstruction method based on harmonic enhancement using the fundamental frequency and phase distortion feature. This feature is known to show fluctuations in the phase spectrum with respect to time and frequency. We estimate the speech phase spectrum by considering the relationship between harmonic phase spectra. Experimental evaluations indicate that the proposed phase reconstruction method improves speech quality in various noisy environments.

13 January 2022 in Advanced Robotics
Verified
Journal Article
Open access
TFTakahiro Fukumori
Takahiro Fukumori
CCChengkai Cai
Chengkai Cai
YZYutao Zhang
Yutao Zhang
LHLotfi El Hafi
Lotfi El Hafi
YHYoshinobu Hagiwara
Yoshinobu Hagiwara
TTTadahiro Taniguchi
Tadahiro Taniguchi
Domestic robots are often required to understand spoken commands in noisy environments, including service appliances' operating sounds. Most conventional domestic robots use electret condenser microphones (ECMs) to record the sound. However, the ECMs are known to be sensitive to the noise in the direction of sound arrival. The laser Doppler vibrometer (LDV), which has been widely used in the research field of measurement, has the potential to work as a new speech-input device to solve this problem. The aim of this paper is to investigate the effectiveness of using the LDV as an optical laser microphone for human-robot interaction in extremely noisy service environments. Our robot irradiates an object near a speaker with a laser and measures the vibration of the object to record the sound. We conducted three experiments to assess the performance of speech recognition using the optical laser microphone in various settings and showed stable performance in extremely noisy conditions compared with a conventional ECM. GRAPHICAL ABSTRACT

1 October 2013
Verified
Conference Paper
Closed access
TFTakahiro Fukumori
Takahiro Fukumori
MNMasato Nakayama
Masato Nakayama
TNTakanobu Nishiura
Takanobu Nishiura
YYYoichi Yamashita
Yoichi Yamashita
The automatic speech recognition (ASR) performance is degraded in noisy and reverberant environments. Although various techniques against degradation of the ASR performance have been proposed, it is difficult to properly apply them in evaluation environments with unknown noisy and reverberant conditions. It is possible to properly apply these techniques for improving the ASR performance if we can estimate the relationship between the ASR performance and degradation factors including both noise and reverberation. In this study, we here propose new noisy and reverberant criteria which are referred as “Noisy and Reverberant Speech Recognition with the PESQ and the D n (NRSR-PD n )”. We first designed the “NRSR-PD n ” using the relationships among the D value, the PESQ score, and the ASR performance. We then estimated the ASR performance with the designed criteria “NRSR-PD n ” in evaluation experiments. Experimental evaluations demonstrated that our proposed criteria make the well suited for robustly estimating the ASR performance in noisy and reverberant environments.

1 April 2012 in The Journal of the Acoustical Society of America
Verified
Journal Article
Closed access
KHKeisuke Horii
Keisuke Horii
TFTakahiro Fukumori
Takahiro Fukumori
MMMasanori Morise
Masanori Morise
TNTakanobu Nishiura
Takanobu Nishiura
YYYoichi Yamashita
Yoichi Yamashita
We should reduce the unwanted noise from noisy speech. Spectral Subtraction (SS) which is one of the noise reduction methods has been proposed by S. F. Boll in 1979, and it can effectively reduce the unwanted noise by utilizing only the observed signal. SS however has a problem that dissonant noise called musical tone is generated after noise reduction. SS estimates the noise with non-speech part and subtracts the estimated noise from observed signal. Flooring process is also performed depend on estimated noise power for supporting the excessive subtraction. Since the musical tone is generated by it, the improvement of SS is required to reduce it. In the past, we proposed SS with the weighted subtraction coefficients in each frequency band for controlling flooring process. In this method, the equal-weighted subtraction coefficients were utilized in every frame, however speech and noise power are different in each frame. To overcome this problem, we newly propose an advanced SS with the color-weighted subtraction coefficients in each frame for effectively reducing the musical tone. Both objective and subjective experiments were carried out for verifying the effectiveness of the proposed SS. As a result, the proposed SS could subjectively reduce the musical tone.

2 October 2024 in IEEE/ACM Transactions on Audio, Speech, and Language Processing
Verified
Journal Article
Open access
TFTakahiro Fukumori
Takahiro Fukumori
TITaito Ishida
Taito Ishida
YYYoichi Yamashita
Yoichi Yamashita
The detection of shouted speech is crucial in audio surveillance and monitoring. Although it is desirable for a security system to be able to identify emergencies, existing corpora provide only a binary label (i.e., shouted or normal) for each speech sample, making it difficult to predict the shout intensity. Furthermore, most corpora comprise only utterances typical of hazardous situations, meaning that classifiers cannot learn to discriminate such utterances from shouts typical of less hazardous situations such as cheers. Thus, this paper presents a novel research source, the RItsumeikan Shout Corpus (RISC), which contains wide variety types of shouted speech samples collected in recording experiments. Each shouted speech sample in RISC has a shout type and is also assigned shout intensity ratings via a crowdsourcing service. We also present a comprehensive performance comparison among deep learning approaches for speech type classification tasks and a shout intensity prediction task. The results show that feature learning based on the spectral and cepstral domains achieves high performance, no matter which network architecture is used. The results also demonstrate that shout type classification and intensity prediction are still challenging tasks, and RISC is expected to contribute to further development in this research area.

1 May 2013 in The Journal of the Acoustical Society of America
Verified
Journal Article
Closed access
PDF file
TFTakayuki Furoh
Takayuki Furoh
TFTakahiro Fukumori
Takahiro Fukumori
MNMasato Nakayama
Masato Nakayama
TNTakanobu Nishiura
Takanobu Nishiura
In noisy environments, the recorded speech is distorted by the additional noise and the Lombard effect. Thus, the automatic speech recognition (ASR) performance is degraded in noisy environments. To solve this problem, noise reduction methods have been proposed as the conventional study. However, in the conventional study, the improvement of ASR performance for the Lombard effect was not discussed well enough. In the present paper, we focus on the robustly detection for Lombard effect speech (Lombard speech). This is because the ASR system can employ a suitable acoustic model by detecting the Lombard speech. We previously proposed the detection for Lombard speech based on second-order MFCC and fundamental frequency. The previously proposed method however requires longer utterances to detect Lombard speech. We therefore newly propose the detection method for Lombard speech with second-order MFCC and spectral envelope in beginning of talking-speech. To detect the Lombard speech at a short time, the proposed method employs variable weights corresponding to elapsed time for second-order MFCC and spectral envelope. As a result of evaluation experiments, we confirmed that the detection time was reduced from the conventional method.