A method for building an acoustic model and a speech decoding method based on the model

Abstract

The present invention provides a method for establishing an acoustic model and a speech decoding method based on the model. The method includes: step 101) calculating the statistics of the three-factor Gaussian mixture model required for the acoustic model based on the training data; step 102) using The self-clustering method calculates the decision tree problem set according to the statistics, uses the decision tree algorithm to segment and cluster the statistics based on the obtained decision tree problem set, and then obtains the isotopic phoneme; step 103) combines the basic phoneme set with the described The allophones are merged as phoneme sets containing allophones, and the original voice annotation is processed through the decision tree process, and the processed voice annotation is called phoneme annotation containing allophones; step 104) based on the phoneme set containing allophones and voice annotation , using an acoustic model training method to train the acoustic model to generate an acoustic model containing isotopes. The present invention will be dedicated to solving the problem of high acoustic model confusion in the agglutinative language speech recognition system.

Description

technical field [0001] The invention relates to the field of speech recognition, and is mainly aimed at an agglutinative speech recognition system. Background technique [0002] In language morphology, according to whether the language needs to rely on the change of word endings to express its grammatical relationship, it can be divided into analytic language and synthetic language. Classification. Adhesive language is a kind of synthetic language, which belongs to the comprehensive language with high inflection. Its word-level units are usually composed of a large number of morpheme connections, which are called agglutinative characteristics. Since the speech recognition system was originally designed for analytic languages ​​and quasi-analytic languages, such as Chinese and English, the emergence of sticky features has brought many new problems to the traditional speech recognition system, which requires further improvement and improvement. improved. [0003] The proble...

AI Technical Summary

Problems solved by technology

However, the number of word-level units in agglutinative language has increased dramatically due to the existence of agglutinative properties, and the number of common words has far exceeded the size that the dictionary can accommodate, so it is not suitable as the basic modeling unit of the language model; at the same time, the secondary natural language unit phoneme (or word , depending on the language and the sub-units are different) is not suitable as the basic modeling unit of the language model, because the sticky characteristics will aggravate the short-sighted phenomenon of this level unit

Second, in terms of acoustic models, the cohesion of phonemes in agglutinative languages ​​will lead to a large number of co-articulation phenomena, that is, the same phoneme will have many different pronunciations depending on its location.

But the second problem has not yet found an effective solution, which is one of the difficulties that plague the speech recognition of agglutinative language

Images