Research Article
Regional Language Speech Recognition from Bone-Conducted Speech Signals through Different Deep Learning Architectures
Table 3
Voice and BCM signal correlation with LSSVM, SVM, and SVR.
| | Correlation between voice and BCM signal | | Tamil words and syllabi | Architecture | Correlation algorithm (%) | | S-Net | UNet | CapsuleNet |
| | Amam (2 syllabi) | 83.29 | 85.62 | 87.52 | LSSVM | | Vena (2 syllabi) | 81.56 | 84.58 | 88.15 | | Iruku (2 syllabi) | 82.69 | 87.56 | 92.15 | | Illa (2 syllabi) | 83.59 | 87.91 | 93.45 | | Enna (2 syllabi) | 83.15 | 88.15 | 91.25 | | Engae (2 syllabi) | 82.18 | 88.94 | 93.58 | | Naan (2 syllabi) | 83.29 | 89.18 | 94.59 | | Va (1 syllabi) | 92.6 | 93.25 | 96.12 | SVM | | Engae va (3 syllabi) | 91.2 | 92.89 | 97.25 | SVR |
|
|
