Speech recognition system, speech recognition method and computer program product

Abstract

The invention discloses a speech recognition system and method thereof, and a computer program product. The speech recognition system is connected to an external general-purpose speech recognition system, and includes a storage unit and a processing unit. The storage unit stores a specific application speech recognition module, a comparison module and an enhancement module. The specific application speech recognition module converts an input speech signal into a first phonetic text. The general-purpose speech recognition system converts the speech signal into a written text. The comparison module receives the first phonetic text from the specific application speech recognition module and the written text from the general-purpose speech recognition system, converts the written text into a second phonetic text, and aligns the second phonetic text with the first phonetic text based on similarity of pronunciation to output a phonetic text alignment result. The enhancement module receives the phonetic text alignment result from the comparison module, and constructs the phonetic text alignment result after a path weighting with the written text and the first phonetic text to form an outputting recognized text.

Description

technical field [0001] The present invention relates to technologies related to speech recognition, in particular to a speech recognition system and its method for a specific application situation, and a computer program product. Background technique [0002] In order to convert speech into text data, a speech recognition system is generally selected as a tool for converting the user's speech information into text data. The common speech recognition system is a general-purpose speech recognition system, such as the Google speech recognition system. Voice information can be converted into text only through a general-purpose voice recognition system, so that chat information can be transmitted on communication software or compiled into articles for public viewing on social networking sites. According to this, users do not need to input word by word with the keyboard. Visible Convenience of speech recognition system. In addition, with the development of smart phones, users can...

AI Technical Summary

Problems solved by technology

However, general-purpose speech recognition systems often only provide text recognition results, and do not provide sufficient operation options and detailed information for developers or users to perform subsequent reinforcement processing, and general-purpose speech recognition systems can only output written text, and The written text often does not have detailed information such as word segmentation and vocabulary confidence, and general-purpose speech recognition systems usually belong to cloud services, and the additional information that users can obtain is usually limited

Therefore, under the general-purpose speech recognition system, it will be difficult for users to improve the situation of inaccurate recognition, especially in specific application scenarios

[0004] As can be seen from the above, under the use of the existing speech recognition system, how to solve the relatively inaccurate recognition situation that may occur in a specific application situation will become a technical challenge that people in the technical field are eager to solve.

Images