摘要 |
Generating words and/or names, comprising: receiving at least one corpus based on a given language (402); generating a plurality of N-grams of phonemes and a plurality of frequencies of occurrence using the corpus (404), such that each frequency of occurrence corresponds to a respective pair of phonemes and indicates the frequency of the second phoneme in the pair following the first phoneme in the pair; generating a phoneme tree (404) using the plurality of N-grams of phonemes and the plurality of frequencies of occurrence; performing a random walk on the phoneme tree (406) using the frequencies of occurrence to generate a sequence of phonemes; and mapping (416) the sequence of phonemes into a sequence of graphemes. |