| 主权项 |
1. A speech recognition method, adapted to an electronic apparatus, comprising:
obtaining a phonetic transcription sequence of a speech signal according to an acoustic model; obtaining a plurality of possible syllable sequences and a plurality of corresponding phonetic spelling matching probabilities according to the phonetic transcription sequence and a syllable acoustic lexicon; obtaining an intonation information corresponding to each of the syllable sequences according to a tone of the phonetic transcription sequence; obtaining a plurality of phonetic spelling sequences and a plurality of phonetic spelling sequence probabilities, from the language model, according to each phonetic spelling of phonetic spelling sequences and the intonation information; obtaining, from the language model, a plurality of text sequences corresponding to the phonetic transcription sequence, and a plurality of spelling sequence probabilities; generating a plurality of associated probabilities by multiplying each of the phonetic spelling matching probabilities and each of the spelling sequence probabilities; and selecting the text sequence corresponding to a largest one among the associated probabilities to be used as a recognition result of the speech signal, wherein different intonation information in the language model is divided into different semantemes, and the semantemes are corresponding to different phonetic spelling sequences. |
