An Amharic speech corpus for large vocabulary continuous speech recognition
An Amharic speech corpus for large vocabulary continuous speech recognition
复制标题
用于大词汇量连续语音识别的阿姆哈拉语语音语料库
DOI:
10.21437/interspeech.2005-467
复制
发表时间:
2005
期刊:
影响因子:
--
通讯作者:
Bairu Tafila
中科院分区:
文献类型:
--
作者:
S. Abate;W. Menzel;Bairu Tafila
• has rich morphology -> many word forms. Phonetics Amharic has a set of 38 phones, seven vowels and thirty-one consonants. Consonants Manner Voicing Place of Articulation of Art/n Lab Dent Pal Vel Glo Stops Voiceless p[p] t[t] m[t∫ ] k[k] [?] Voiced b[b] d[d] ¥[d ] g[g] GlottalizedI[p‘] μ[t‘] 1⁄2[t∫ ‘]q[q] Rounded [kw], [gw], [qw] Fricatives Voiceless f[f] s[s] ][∫ ] h[h] Voiced z[z] [ ] Glottalized O[s‘] Rounded [hw] Nasals Voiced m[m]n[n] }[ ] Liquids Voiced l[l], r[r] Semi vowelsVoiced w[w] y[j]