A Russian Continuous Speech Recognition System Based on the DTW Algorithm under Artificial Intelligence

Yu, Chunping; Wang, Xin

doi:10.1155/2022/5777472

Cited by 3 publications

(2 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2C). The objective function of Bayesian optimization is defined as the similarity between the synthesized and actual sentences and is scaled by custom dynamic-time-warping (DTW) distance that has been selected to evaluate the similarity between two sequence data having potential common features with different length (37)(38)(39)(40). Bayesian optimization gives the optimal segmentation parameters by minimizing the defined DTW distance.…”

Section: Artificial Synthesis Of Natural Fluent Lip Speechmentioning

confidence: 99%

“…The objective function is defined as the similarity between the synthesized and actual sentences and is scaled by custom DTW distance. DTW gives the minimal cumulative distance between two sequences with different lengths, which represents their similarity (38)(39)(40). Here, the synthesized sentence S syn and the actual sentence S act are two sequences with four channels (rows) and different lengths (columns, n and m, respectively).…”

Section: Bayesian Optimization For Segmentation Parameter Recognitionmentioning

confidence: 99%

See 1 more Smart Citation

A data-efficient and easy-to-use lip language interface based on wearable motion capture and speech movement reconstruction

Liu,

Fawden,

Zhu

et al. 2024

Sci. Adv.

View full text Add to dashboard Cite

Lip language recognition urgently needs wearable and easy-to-use interfaces for interference-free and high-fidelity lip-reading acquisition and to develop accompanying data-efficient decoder-modeling methods. Existing solutions suffer from unreliable lip reading, are data hungry, and exhibit poor generalization. Here, we propose a wearable lip language decoding technology that enables interference-free and high-fidelity acquisition of lip movements and data-efficient recognition of fluent lip language based on wearable motion capture and continuous lip speech movement reconstruction. The method allows us to artificially generate any wanted continuous speech datasets from a very limited corpus of word samples from users. By using these artificial datasets to train the decoder, we achieve an average accuracy of 92.0% across individuals ( n = 7) for actual continuous and fluent lip speech recognition for 93 English sentences, even observing no training burn on users because all training datasets are artificially generated. Our method greatly minimizes users’ training/learning load and presents a data-efficient and easy-to-use paradigm for lip language recognition.

show abstract

Section: Artificial Synthesis Of Natural Fluent Lip Speechmentioning

confidence: 99%

Section: Bayesian Optimization For Segmentation Parameter Recognitionmentioning

confidence: 99%

A data-efficient and easy-to-use lip language interface based on wearable motion capture and speech movement reconstruction

Liu,

Fawden,

Zhu

et al. 2024

Sci. Adv.

View full text Add to dashboard Cite

show abstract

Retracted: A Russian Continuous Speech Recognition System Based on the DTW Algorithm under Artificial Intelligence

2023

Journal of Robotics

View full text Add to dashboard Cite

Technical Application of Artificial Intelligence-Based Voice Interaction Control and Intelligent Scheduling for Throwing and Retiring Reclosing Gates

Bai,

Lin,

Peng

et al. 2024

Applied Mathematics and Nonlinear Sciences

View full text Add to dashboard Cite

Speech recognition, as an important carrier of human information exchange, is characterized by ease of use and high efficiency, and is an excellent choice to act as a medium of information interaction between humans and machines. In this paper, with the background of the application of throwback reclosing in the field of electric power, a speech recognition model is constructed using the DTW algorithm. Subsequently, on this basis, an intelligent dispatching command generation system integrating speech recognition, generation, and interaction systems was built, and the functional characteristics of the intelligent reclosing model were characterized through simulation experiments. After 40-50 rounds of training, the Loss curve of the model in this paper settles to about 0.16. After training for 100 rounds, the word error rate stabilizes at around 7%. In the call corpus, the speech recognition system constructed in this paper has a error rate of 1.124%. The recall in the Scheduling instruction corpus and Scheduling telephone corpus is 0.982 and 0.947 respectively. And, the model in this paper has better performance in terms of the number of errors in speech recognition samples and the word error rate of specialized vocabulary. In addition, the model in this paper can normalize the correlation of electricity consumption between different time points under abnormal conditions. In conclusion, the model in this paper is both efficient and stable in the power field, especially in the application of reclosing technology.

show abstract

A Russian Continuous Speech Recognition System Based on the DTW Algorithm under Artificial Intelligence

Cited by 3 publications

References 17 publications

A data-efficient and easy-to-use lip language interface based on wearable motion capture and speech movement reconstruction

A data-efficient and easy-to-use lip language interface based on wearable motion capture and speech movement reconstruction

Retracted: A Russian Continuous Speech Recognition System Based on the DTW Algorithm under Artificial Intelligence

Technical Application of Artificial Intelligence-Based Voice Interaction Control and Intelligent Scheduling for Throwing and Retiring Reclosing Gates

Contact Info

Product

Resources

About