A method and system for automatic speech recognition based on artificial intelligence
A technology of automatic speech recognition and artificial intelligence, applied in speech recognition, speech analysis, instruments, etc., can solve problems such as low efficiency of professional vocabulary recognition and inaccurate recognition of professional vocabulary, so as to reduce professional misunderstandings, improve professionalism, and improve search speed effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0081] Embodiments of the present invention provide a method for automatic speech recognition based on artificial intelligence, figure 1 It is a flow chart of an artificial intelligence-based automatic speech recognition method in an embodiment of the present invention, please refer to figure 1 , the method includes the following steps:
[0082] Step S101, receiving a voice signal to be recognized;
[0083] Step S102, preprocessing the speech signal to be recognized to obtain a speech input signal;
[0084] Step S103, converting the voice input signal from the time domain to the frequency domain, and extracting voice feature parameters;
[0085] Step S104, randomly sampling the speech characteristic parameters to obtain several sample characteristic parameters;
[0086] Step S105, input the sample feature parameters into the acoustic model and the language model, and obtain the recognition result through decoding and searching;
[0087] Step S106, input the recognition res...
Embodiment 2
[0098] On the basis of Embodiment 1, after describing and outputting the corresponding text, it includes:
[0099] Inputting the outputted text into the spelling error correction model to obtain the text after error correction;
[0100] Output the corrected text as the final text.
[0101] The working principle and beneficial effects of the above technical solution are as follows: the solution adopted in this embodiment is a process of spelling error correction for the input text. After passing through the acoustic model and the language model, the output text may have spelling errors and other forms. The problem is that in order to ensure the accuracy and professionalism of automatic speech recognition, it is necessary to correct the spelling of the output text later. By setting the spelling error correction model, it is ensured that the final output text has no formal spelling errors and the accuracy of automatic speech recognition is improved. .
Embodiment 3
[0103] On the basis of embodiment 1, described vocabulary classification template construction method comprises:
[0104] Acquire a large number of professional vocabulary belonging to different industries;
[0105] The professional vocabulary is classified and trained according to the industry to which the professional vocabulary belongs using a convolutional neural network;
[0106] A classification result is obtained, and the classification result is stored in a classification database to form a vocabulary classification template.
[0107]The working principle of the above technical solution is: the solution adopted in this embodiment is a description of the construction method of the vocabulary classification template. By obtaining a large number of professional vocabulary in different industries, the convolutional neural network is used to classify and train the above-mentioned professional vocabulary based on the industry. Classify, and store the classification results...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More