Method and device for identifying dialect types

Abstract

The invention provides a method and device for identifying dialect types. The method comprises the steps of: acquiring a first acoustic feature and a second acoustic feature of each syllable in a to-be-detected syllable sequence; inputting the first acoustic feature of each syllable to a trained pronouncing type model to obtain the pronouncing type of each syllable; inputting the pronouncing typesto a trained pronouncing type combination model to obtain a first probability; inputting the second acoustic feature of each syllable to a trained tone model to obtain a second probability; determining the dialect type to which the syllable sequence belongs according to the product of the first probability and the second probability. The method and the device solve the technical problems of low accuracy and poor universal applicability of conventional dialect identification methods.

Description

technical field [0001] The invention relates to the field of speech signal processing, in particular to a method and device for identifying dialect types. Background technique [0002] Voice is one of the most direct, convenient and natural ways for human beings to realize information interaction. Researchers have been working for years to make computers and even robots use speech to interact like natural people. Although the fields of speech synthesis, speech recognition and natural language processing have made great progress and are widely used, the research on language interaction based on multiple Chinese dialects is very rare. Among them, the research on Chinese dialect recognition will be an important part of many intelligent speech technologies, which can provide guarantee and improve efficiency for the next step of information processing. [0003] The existing research on Chinese dialect identification is still in its infancy. Most studies regard Chinese dialect i...

AI Technical Summary

Problems solved by technology

[0003] The existing research on Chinese dialect identification is still in its infancy. Most studies regard Chinese dialect identification as a special case of language identification, and simply replace the research object with Chinese dialects on the basis of existing language identification methods. , so the recognition accuracy is low

In addition, there are many existing Chinese dialects, and there are many divisions within the dialects. However, the Chinese dialect identification methods in the prior art are to develop an identification method for a dialect, and there is no universal identification as a system. method

[0004] In addition, although there are studies on initials and finals in the prior art, they are all based on the division and utilization of the surface information of initials and finals. The corpus data in the existing Chinese dialect recognition methods are often very single, and the high-level linguistic information It is also rarely used, resulting in low accuracy of dialect identification and not universally applicable

[0005] Aiming at the problems of low accuracy and not universal applicability in the above-mentioned dialect identification method in the prior art, no effective solution has been proposed yet

Images