Bioinspired dual-channel speech recognition using graphene-based electromyographic and mechanical sensors
Source abstract
Abstract
Automatic speech recognition (ASR) is helpful to improve quality of life. However, the performance of ASR degrades in the case of noisy environment, limited privacy, and speech disorders. Herein, we analyze the generation mechanism of speech and utilize dual-biological-channel for speech recognition. We also propose a dual-channel graphene-based electromyographic (EMG) and mechanical sensor (DGEMS) that can simultaneously collect two bio-signals at the same point. The EMG electrodes have better performance compared with commercial electrodes, and the mechanical sensors exhibit excellent repeatability during 10 million fatigue testing. Based on the excellent performance of EMG electrodes and mechanical sensors, 100% accuracy is achieved on digits dataset and 96.85% accuracy on a dataset containing 71 words. We also demonstrate that the DGEMS can be very resistant to noise, and the accuracy is always higher than 95%. Dual biological channels can greatly improve the performance of speech recognition in more scenarios.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Muscle activation and electromyography studies · Advanced Sensor and Energy Harvesting Materials · EEG and Brain-Computer Interfaces
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (35)
Show source-provided references
1. Improving noise robustness of automatic speech recognition via parallel data and teacher-student learning
2. Robust speech recognition via anchor word representations
3. Sign-to-speech translation using machine-learning-assisted stretchable sensor arrays10.1038/s41928-020-0428-6 ↗
4. Alterego: a personalized wearable silent speech interface
5. Gesture recognition using a bioinspired learning architecture that integrates visual data with somatosensory data from stretchable sensors10.1038/s41928-020-0422-z ↗
6. State-of-the-art speech recognition using eeg and towards decoding of speech spectrum from eeg
7. Speech synthesis from neural decoding of spoken sentences10.1038/s41586-019-1119-1 ↗
8. Speech synthesis from ECoG using densely connected 3D convolutional neural networks10.1088/1741-2552/ab0c59 ↗
9. A bioinspired, durable, and nondisposable transparent graphene skin electrode for electrophysiological signal detection10.1021/acsmaterialslett.0c00203 ↗
10. A wearable skinlike ultra-sensitive artificial graphene throat10.1021/acsnano.9b03218 ↗
11. Speaker-independent silent speech recognition from flesh-point articulatory movements using an LSTM neural network10.1109/TASLP.2017.2758999 ↗
12. Digital voicing of silent speech
13. Development of sEMG sensors and algorithms for silent speech recognition10.1088/1741-2552/aac965 ↗
14. Direct conversion from facial myoelectric signals to speech using deep neural networks
15. An intelligent artificial throat with sound-sensing ability based on laser induced graphene10.1038/ncomms14579 ↗
16. Ultrasensitive anti-interference voice recognition by bio-inspired skin-attachable self-cleaning acoustic sensors10.1021/acsnano.9b06354 ↗
17. A durable nanomesh on-skin strain gauge for natural skin motion monitoring with minimum mechanical constraints10.1126/sciadv.abb7043 ↗
18. Wand · 2015
19. Skeletal muscle physiology
20. Linear elastic properties of the facial soft tissues using an aspiration device: towards patient specific characterization10.1007/s10439-014-1098-1 ↗
21. Image-based skeletal muscle coordination: case study on a subject specific facial mimic simulation10.1142/S0219519418500203 ↗
22. A universal method to grow and etch graphene film
23. Speech recognition for vocalized and subvocal modes of production using surface EMG signals from the neck and face
24. Silent speech recognition based on surface electromyography
25. Feature selection of mime speech recognition using surface electromyography data
26. Inductive conformal prediction for silent speech recognition10.1088/1741-2552/ab7ba0 ↗
27. Deep-learning-enabled MXene-based artificial throat: toward sound detection and speech recognition10.1002/admt.202000262 ↗
28. Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network10.1038/s41591-018-0268-3 ↗
29. A computationally efficient sEMG based silent speech interface using channel reduction and decision tree based classification10.1016/j.procs.2020.04.013 ↗
30. Attention bidirectional LSTM networks based mime speech recognition using sEMG data
31. Non-invasive silent speech recognition in multiple sclerosis with dysphonia
32. Comparison of feature evaluation criteria for speech recognition based on electromyography10.1007/s11517-017-1723-x ↗
33. EMG-based speech recognition using dimensionality reduction methods
34. Bioinspired in-sensor visual adaptation for accurate perception10.1038/s41928-022-00713-1 ↗
35. Near-sensor and in-sensor computing10.1038/s41928-020-00501-9 ↗