Language identification and classification method and device based on noise reduction automatic encoder

Abstract

The invention provides a language identification and classification method based on a noise reduction automatic encoder. The method includes: step 1) extracting a to-be-identified speech signal from ato-be-identified speech segment to obtain underlying acoustic characteristics; step 2) extracting an original i-vector from the underlying acoustic characteristics obtained in step 1); step 3) calculating and obtaining a phoneme vector pc (u); step 4) splicing the original i-vector with the phoneme vector pc (u), and then inputting to a DAE-based i-vector compensation network to obtain a compensated i-vector; step 5) inputting the original i-vector obtained in step 2) and the compensated i-vector obtained in step 4) respectively into a pre-trained logistic regression classifier to obtain corresponding score vectors; and step 6) performing score fusion on the corresponding score vectors obtained in step 5) to obtain a final score vector, further obtaining a probability of each language category, and determining the language category to which it belongs.

Description

technical field [0001] The invention belongs to the technical field of language recognition, and in particular relates to a method and device for language recognition and classification based on a noise reduction automatic encoder. Background technique [0002] Language Identification (LID) refers to the process of automatically determining a given speech segment, extracting the difference information of each language from the speech signal of the speech segment, and judging the language type. Language recognition technology has important applications in multilingual speech processing, such as spoken language translation systems, multilingual speech recognition systems, speech and text processing, etc. [0003] At present, the traditional language recognition technology includes two methods: the first method is the language recognition technology based on the features of the phoneme layer; among them, the language recognition technology based on the features of the phoneme l...

AI Technical Summary

Problems solved by technology

At present, the existing language recognition system has a high recognition rate when the length of the training speech and the test speech match; however, when the length of the training speech and the test speech do not match, its performance also decreases

The existing language recognition system, for the length mismatch problem, trains the matching models for different lengths of test speech, which greatly increases the complexity of the model

Images