Sign Language Recognition using VGG16 and ResNet50 DOI
Rohan Gupta,

Krishnam Gupta,

Chirag Pandit

et al.

2022 9th International Conference on Computing for Sustainable Global Development (INDIACom), Journal Year: 2024, Volume and Issue: unknown, P. 996 - 1001

Published: Feb. 28, 2024

Sign language is a vital way of carrying out the conversation among mute and deaf people. The has flaw that it not known to everyone. To overcome this problem, Recognition program comes into picture. proposed model been planned make simpler between persons. For determining movements gestures employs neural network along with several algorithms. used two major networks which are Resnet50 (Residual Network) VGG16 (Visual Geometry Group). trained based on diverse linguistics across world. It enables user-accessible online application; identifying signs in real time conversion images text few key features model. In education system people study content without making any extra effort. can also be great use healthcare could work as an effective means communication patients doctors, would improve care quality patients. observed were recognized more accurately help using both networks. was provided sample divided folders i.e., test train 70 percent transferred folder whereas 30 after went through testing desired outcome identification achieved.

Language: Английский

Comparative study of low resource Digaru language using SMT and NMT DOI
Rushanti Kri, Koj Sambyo

International Journal of Information Technology, Journal Year: 2024, Volume and Issue: 16(4), P. 2015 - 2024

Published: March 7, 2024

Language: Английский

Citations

3
Spoken word recognition using a novel speech boundary segment of voiceless articulatory consonants DOI
Bachchu Paul, Sumita Guchhait,

Sandipan Maity

et al.

International Journal of Information Technology, Journal Year: 2024, Volume and Issue: 16(4), P. 2661 - 2673

Published: March 17, 2024

Language: Английский

Citations

2
Acoustic and Spectral Analysis of Adi Triphthongs DOI
Sajal Sasmal, Yang Saring

IETE Journal of Research, Journal Year: 2024, Volume and Issue: 70(8), P. 6572 - 6582

Published: Feb. 12, 2024

Adi, a zero-resource indigenous tribal language of Arunachal Pradesh, originated in Tibeto-Burman. According to the "UNESCO Atlas World's Languages Danger 2017" report, Adi is critically endangered. In this research, two new triphthongs, /uai/ [uai] and /oai/ [ɔai], are proposed phoneme set language. has 51 phonemes, including 14 monophthongs (7 short 7 long vowels), 19 diphthongs, 16 consonants. This research analysed triphthongs' spectral acoustic properties, intensity variation, long-term average spectrum (LTAS), formant distribution. The mean first four frequencies male female speakers measured, which may be identification these phones speech recognition systems. Four different words for each triphthong recorded from 68 individual native (32 males 36 females) Pradesh. study helpful phonetic modelling build applications

Language: Английский

Citations

1
Sign Language Recognition using VGG16 and ResNet50 DOI
Rohan Gupta,

Krishnam Gupta,

Chirag Pandit

et al.

2022 9th International Conference on Computing for Sustainable Global Development (INDIACom), Journal Year: 2024, Volume and Issue: unknown, P. 996 - 1001

Published: Feb. 28, 2024

Sign language is a vital way of carrying out the conversation among mute and deaf people. The has flaw that it not known to everyone. To overcome this problem, Recognition program comes into picture. proposed model been planned make simpler between persons. For determining movements gestures employs neural network along with several algorithms. used two major networks which are Resnet50 (Residual Network) VGG16 (Visual Geometry Group). trained based on diverse linguistics across world. It enables user-accessible online application; identifying signs in real time conversion images text few key features model. In education system people study content without making any extra effort. can also be great use healthcare could work as an effective means communication patients doctors, would improve care quality patients. observed were recognized more accurately help using both networks. was provided sample divided folders i.e., test train 70 percent transferred folder whereas 30 after went through testing desired outcome identification achieved.

Language: Английский

Citations

1