| 摘要 |
1,224,137. Speech synthesizer. NIPPON TELEGRAPH & TELEPHONE PUBLIC CORP. 30 April, 1969 [1 May, 1968], No. 21992/69. Heading H4R. Synthetic speech is built up by the concatenation of phonemic units each of which begins and ends with a vowel, e.g. a vowel, consonant, vowel unit or a vowel-vowel unit. Since the acoustic features of a vowel vary relatively slowly phonemic units may more easily be joined in the mid portion of a vowel. For a word which begins or ends with a consonant a phoneme unit which comprises the same vowel before and after the required consonant is used and the redundant vowel is removed. Nasal sounds may also be considered as vowels and appear at the beginning and/or end of a phoneme unit. The apparatus may include stored information to modify the stored phoneme units to produce the effect of different accents or talkers, e.g. male and female voices. |
