Syllable-based recognition of Arabic & English digits in noisy environment

The presence of noise degrades the recognition percent of automatic speech recognition systems. In this paper, we concentrate on the comparison between Arabic and English digits in different conditions of noise using different acoustic units. Speaker-independent hidden Markov models (HMMs)-based speech recognition system was designed using Hidden Markov model toolkit (HTK). The database used of Arabic consists from seventy-five Egyptian speakers. But the database of English is the AURORA Corpus. In Arabic database, experiments show that the recognition rate using syllables outperformed the rate obtained using monophones, triphones and words in different conditions of noise. But in Aurora database, recognition using words or syllables outperforms monophones and triphones. Recognition using words are very close to using syllables because all tidigits of Aurora corpus are monosyllabic Moreover, experiments show that speech recognition using syllables is more robustness to noise than triphones and monophones.