Effect of characteristics of speakers on MSA ASR performance

Droua-Hamdani, Ghania; Sellouani, S.; Boudraa, Malika

doi:10.1109/iccspa.2013.6487262

Cited by 10 publications

(2 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Clean data were decoded using Kaldi decoder. The evaluation of accuracy needs to be done using word error rate (WER).The Word error rate [14] is calculated based on the clean data and were decoded using Kaldi decoder. The evaluation of accuracy is done using word error rate (WER).…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Automatic speech recognition based Odia system

Karan

Sahoo

Sahu

2015

2015 International Conference on Microwave, Optical and Communication Engineering (ICMOCE)

View full text Add to dashboard Cite

In this paper, we presents a scheme of implementation of automatic speech recognition system using Odia language. The proposed system is realized using Kaldi toolkit for automatic recognition of district models of the state of Odisha. We investigate the monophone and triphone model for Odia speech recognition. Acoustic modeling of Odia speech recognition was performed using HMM and GMM. A significant reduction in word error rate (WER) was observed using the triphone model. We report the preliminary speech recognition results using Kaldi speech toolkit.

show abstract

Section: Resultsmentioning

confidence: 99%

“…The evaluation of accuracy is done using word error rate (WER). The Word error rate [14] is calculated as WER = [(S + D + I) / N]* 100 (1) Where S= Number of substitution D= Number of deletion I= Number of insertion N= Number of words in reference…”

Section: Resultsmentioning

confidence: 99%