Speech Recognition On Fongbe Speech
評価指標
Word Error Rate (WER)
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
モデル名 | Word Error Rate (WER) | Paper Title | Repository |
---|---|---|---|
Triphone (39 features) + LDA and MLLT + SAT and FMLLR | 17.77 | First Automatic Fongbe Continuous Speech Recognition System: Development of Acoustic Models and Language Models | |
Triphone (13 MFCC + delta + delta2) | 26.75 | First Automatic Fongbe Continuous Speech Recognition System: Development of Acoustic Models and Language Models | |
Triphone (39 features) + LDA and MLLT + SGMM | 16.57 | First Automatic Fongbe Continuous Speech Recognition System: Development of Acoustic Models and Language Models |
0 of 3 row(s) selected.