Method for identifying national language single tone and sentence with a hundred percent identification rate

Abstract

The present invention relates to a Mandarin monosyllabic word and sentence recognition method, which comprises: inviting an enunciator to pronounce each monosyllabic word, and obtaining K samples of the monosyllabic word with the shortest Baye distance to the known monosyllabic word of the enunciator in a database (i.e., take K samples with the shortest Baye distance to the known monosyllabic word of the enunciator as the best samples of the monosyllabic word, and then abstract characteristics of the K best samples to represent the monosyllabic word, and store the abstracted characteristics in a database). Since K best samples are available to calculate the characteristics for each monosyllabic word, the monosyllabic word recognition capability of the present invention is enhanced greatly. Next, a database of sentences and names is created for the sentences and names to be recognized. In a test of 390 monosyllabic words and 460 sentences and names pronounced by three male and female persons, the successful recognition ratio is 100 percent. In addition, sentences or names can be added to the database at any time, and then the sentences or names can be recognized with the database. Above all, the present invention provides a method for correcting the characteristics of monosyllabic words to ensure successful recognition.

Description

technical field [0001] What the present invention relates to is a kind of Mandarin single-syllable and sentence recognition method, what also particularly relate to is a kind of in a clear pronunciation person in the database, first find K "best" sample average and variation for each single-syllable The numbers represent the single tones. The E×P=144 characteristic ranges of the single tone are clearly displayed and will not overlap with other single tone ranges. The Bayesian classification method clearly compares the characteristics of the unknown monophony and the characteristics of the known monophony to improve the recognition ability of the present invention. In detail, the speech recognition method of the present invention includes E equal-length elastic frames, without filters, without overlapping, to frame single-tone sound waves of different lengths, and to normalize the sound waves and convert them into E linear predictive coding inversions. Spectrum (LPCC) vector....

AI Technical Summary

Problems solved by technology

Recognition will be good, but pulling the same features to the same position is difficult and warping takes too long to apply

The vector quantization method is not only inaccurate but also time-consuming to identify a large number of single tones

The hidden Markov model method (HMM) is a good identification method recently, but the method is complicated, too many unknown parameters need to be estimated, and it takes time to calculate the estimated value and identify

Recently, T.F.Li used the Bayesian classification method in the paper Speech recognition of mandarin monosyllables published in Pattern Recognition, vol.36 in 2003, and used the same database to compress various long and short sequences of LPCC vectors into classification models of the same size. The result is better than Y.K.Chen, C.Y.Liu, G.H.Chiang, M.T.Lin published in Proceedings of Telecommunication Symposium, Taiwan in 1990, the paper The recognition of mandarin monosyllables based on the discrete hidden Markov model HMM method is better, but the compression The process is complicated and time-consuming, and it is difficult to compress the same features to the same time position for the same single sound, and it is difficult to identify similar single sounds

Images