Deep CNN Based Feature Extractor for Text-Prompted Speaker Recognition

Novoselov, Sergey; Kudashev, Oleg; Shchemelinin, Vadim; Kremnev, Ivan; Lavrentyeva, GaZina

doi:10.1109/icassp.2018.8462358

Cited by 13 publications

(7 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In particular, for stimuli identification and feature analysis in HMI devices, which require training with real data, supervised learning algorithms are the most popular because they ensure fast and efficient performance. For example, Lee et al focused on the integration of machine learning with flexible piezoelectric acoustic sensors, showing huge potential for speech recognition [91] . Figure 6A briefly illustrates the process of machine learning-enhanced P-HMIs and their potential applications in HMIs concerning human voices.…”

Section: Machine Learning-enhanced P-hmismentioning

confidence: 99%

Recent progress in flexible piezoelectric devices toward human-machine interactions

2022

Soft Sci

View full text Add to dashboard Cite

Human-machine interactions are becoming increasingly required for intelligent sensing and effective manipulation. Recent developments in flexible piezoelectric sensors with short response time and high force-electric interconversion efficiency present a tendency toward facilitating diverse human-machine interactive applications. Here, we review the development of flexible piezoelectric human-machine interactions in the context of robotic control, the Internet of Things, sports coaching and acoustic therapeutics. The synthesis of unique materials, the distinct design of device structures, the typical applications of piezoelectric human-machine interactions and the integration of cutting-edge technologies are elaborated in detail based on recent research. Finally, we highlight the current challenges and directions for the development of piezoelectric human-machine interactions for more advanced application scenarios.

show abstract

Section: Machine Learning-enhanced P-hmismentioning

confidence: 99%

Recent progress in flexible piezoelectric devices toward human-machine interactions

2022

Soft Sci

View full text Add to dashboard Cite

show abstract

“…The idea of local perception has been widely applied in the feature-extracted full-channel and local attention channels, used as feature extraction tools to obtain text features [30]. Multi-mode information also has been adopted in the recommendation of microblogging [31], using CNN and RNN to extract features from images and texts, and then combining them to make tag recommendations.…”

Section: Deep Learning Models In Recommendationmentioning

confidence: 99%

CDMF: A Deep Learning Model based on Convolutional and Dense-layer Matrix Factorization for Context-Aware Recommendation

Gan¹,

Ma²,

Xiao³

2019

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

We proposes a novel deep neural network based recommendation model named Convolutional and Dense-layer Matrix Factorization (CDMF) for Context-aware recommendation, which is to combine multi-source information from item description and tag information. CDMF adopts a convolution neural network to extract hidden feature from item description as document and then fuses it with tag information via a full connection layer, thus generates a comprehensive feature vector. Based on the matrix factorization method, CDMF makes rating prediction based on the fused information of both users and items. Experiments on a real dataset show that the proposed deep learning model obviously outperforms the state-of-art recommendation methods.

show abstract

“…Textdependent SV task allows us to compare utterances of the same phonetic context [6], [7], or random word sequences coming from a fixed vocabulary [8], [9]. With random sequences, such as random digit strings, an SV system is less vulnerable to replay attacks [8], [10], [11]. In this work, we study a neural acoustic-phonetic approach for SV of random digit strings in RSR2015 Part III database [12].…”

Section: Introductionmentioning

confidence: 99%

Neural Acoustic-Phonetic Approach for Speaker Verification With Phonetic Attention Mask

Liu

Das²,

Lee

et al. 2022

IEEE Signal Process. Lett.

View full text Add to dashboard Cite

Traditional acoustic-phonetic approach makes use of both spectral and phonetic information when comparing the voice of speakers. While phonetic units are not equally informative, the phonetic context of speech plays an important role in speaker verification (SV). In this paper, we propose a neural acousticphonetic approach that learns to dynamically assign differentiated weights to spectral features for SV. Such differentiated weights form a phonetic attention mask (PAM). The neural acoustic-phonetic framework consists of two training pipelines, one for SV and another for speech recognition. Through the PAM, we leverage the phonetic information for SV. We evaluate the proposed neural acoustic-phonetic framework on the RSR2015 database Part III corpus, that consists of random digit strings. We show that the proposed framework with PAM consistently outperforms baseline with an equal error rate reduction of 13.45% and 10.20% for female and male data, respectively.

show abstract

Deep CNN Based Feature Extractor for Text-Prompted Speaker Recognition

Cited by 13 publications

References 20 publications

Recent progress in flexible piezoelectric devices toward human-machine interactions

Recent progress in flexible piezoelectric devices toward human-machine interactions

CDMF: A Deep Learning Model based on Convolutional and Dense-layer Matrix Factorization for Context-Aware Recommendation

Neural Acoustic-Phonetic Approach for Speaker Verification With Phonetic Attention Mask

Contact Info

Product

Resources

About