Speech to text conversion engine for non-standard speech

Patent metadata

Field	Value
Publication number	US-2020335099-A1
Application number	US-201916385017-A
Country	US
Kind code	A1
Filing date	Apr 16, 2019
Priority date	Apr 16, 2019
Publication date	Oct 22, 2020
Grant date	—

Abstract

Using a computing device to convert verbal communications including non-standard speech to text. The computing device receives an audio recording of voice and generates a standard text log. A standard word dictionary is retrieved. Non-standard words not found in the word dictionary are determined. Portions of the audio recording corresponding to the non-standard words are retrieved. Portions of the audio recording corresponding to non-standard words into input into a natural language understanding model. The computing device utilizes the results of the natural language understanding model to determine a best-match non-standard dictionary. One or more portions of the audio recording are used to generate a non-standard text log. The standard text log and non-standard text log are merged.

First claim

What is claimed is:

1 . A method of using a computing device to convert verbal communications including non-standard speech to text, the method comprising:
receiving, by a computing device, an audio recording of voice and converting the audio recording to a computer-usable form usable by the computing device; analyzing by the computing device the computer-usable form of the audio recording to generate a standard text log for the audio recording; retrieving a standard word dictionary; comparing one or more words from the standard text log to the standard word dictionary to determine one or more non-standard words not found in the word dictionary; retrieving one or more portions of the audio recording corresponding to the one or more non-standard words; inputting the one or more portions of the audio recording corresponding to the one or more non-standard words into a natural language understanding model; utilizing by the computing device results of the natural language understanding model to determine a best-match non-standard dictionary for interpreting the one or more non-standard words; retrieving the best-match non-standard dictionary; analyzing by the computing device the one or more portions of the audio recording corresponding to the one or more non-standard words utilizing the best-match non-standard dictionary to generate a non-standard text log; and merging by the computing device the standard text log and the non-standard text log to generate a complete text log.

2 . The method of claim 1 , wherein converting the audio recording to the computer-usable form comprises:
parsing the audio recording into one or more syllables; and storing by the computing device the one or more syllables into one or more data structures to generate the computer-usable form.

3 . The method of claim 1 , wherein the non-standard dictionary is selected from the group consisting of a business dictionary, a technical dictionary, a medical dictionary, a sound dictionary, and a foreign language dictionary.

4 . The method of claim 1 , wherein the analyzing of the computer-usable form of the audio recording comprises analyzing a wave form of the audio recording.

5 . The method of claim 1 , further comprising displaying the complete text log to a user.

6 . The method of claim 1 , wherein comparing one or more words from the standard text log to the standard word dictionary comprises utilizing domain-specific rules.

7 . The method of claim 1 , wherein the natural language understanding model utilizes, at least in-part, domain-specific rules unique to each non-standard dictionary for interpreting the one or more non-standard words.

8 . The method of claim 1 , further comprising utilizing by the computing device the natural language understanding model to analyze the standard text log and further determine one or more non-standard words in the standard text log.

9 . A computer program product utilizing a computing device to convert verbal communications including non-standard speech to text, the computer program product comprising:
one or more non-transitory computer-readable storage media and program instructions stored on the one or more non-transitory computer-readable storage media, the program instructions, when executed by the computing device, cause the computing device to perform a method comprising:
receiving, by a computing device, an audio recording of voice and converting the audio recording to a computer-usable form usable by the computing device;
analyzing by the computing device the computer-usable form of the audio recording to generate a standard text log for the audio recording;
retrieving a standard word dictionary;
comparing one or more words from the standard text log to the standard word dictionary to determine one or more non-standard words not found in the word dictionary;
retrieving one or more portions of the audio recording corresponding to the one or more non-standard words;
inputting the one or more portions of the audio recording corresponding to the one or more non-standard words into a natural language understanding model;
utilizing by the computing device results of the natural language understanding model to determine a best-match non-standard dictionary for interpreting the one or more non-standard words;
retrieving the best-match non-standard dictionary;
analyzing by the computing device the one or more portions of the audio recording corresponding to the one or more non-standard words utilizing the best-match non-standard dictionary to generate a non-standard text log; and
merging by the computing device the standard text log and the non-standard text log to generate a complete text log.

10 . The computer program product of claim 9 , wherein converting the audio recording to the computer-usable form comprises:
parsing the audio recording into one or more syllables; and storing by the computing device the one or more syllables into one or more data structures to generate the computer-usable form.

11 . The computer program product of claim 9 , wherein the non-standard dictionary is selected from the group consisting of a business dictionary, a technical dictionary, a medical dictionary, a sound dictionary, and a foreign language dictionary.

12 . The computer program product of claim 9 , wherein the analyzing of the computer-usable form of the audio recording comprises analyzing a wave form of the audio recording.

13 . The computer program product of claim 9 , further comprising displaying the complete text log to a user.

14 . The computer program product of claim 9 , wherein comparing the one or more words from the standard text log to the standard word dictionary comprises utilizing domain-specific rules.

15 . The computer program product of claim 9 , wherein the natural language understanding model utilizes, at least in-part, domain-specific rules unique to each non-standard dictionary for interpreting the one or more non-standard words.

16 . The computer program product of claim 9 , further comprising utilizing by the computing device the natural language understanding model to analyze the standard text log and further determine one or more non-standard words in the standard text log.

17 . A computer system utilizing a computing device to convert verbal communications including non-standard speech to text, the computer system comprising:
one or more computer processors; one or more computer-readable storage media;
program instructions stored on the computer-readable storage media for execution by at least one of the one or more processors, the program instructions comprising:
program instructions to receive, by a computing device, an audio recording of voice and converting the audio recording to a computer-usable form usable by the computing device;
program instructions to analyze by the computing device the computer-usable form of the audio recording to generate a standard text log for the audio recording;
program instructions to retrieve a standard word dictionary;
program instructions to compare one or more words from the standard text log to the standard word dictionary to determine one or more non-standard words not found in the word dictionary;
program instructions retrieve one or more portions of the audio recording corresponding to the one or more non-standard words;
program instructions to input the one or more portions of the audio recording corresponding to the one or more non-standard words into a natural language understanding model;
program instructions to utilize by the compu

Assignees

• IBM

Inventors

• Ponniah Anbumunee
• Thankappan Nair Satheesh Kumar
• Mathew Nithin
• Rajoriya Ashish
• Malgawa Ashish
• Garg Mansi

Classifications

• G10L2015/027

  Syllables being the recognition units · CPC title

• G10L15/18

  using natural language modelling · CPC title

• G10L15/02 Primary

  Feature extraction for speech recognition; Selection of recognition unit · CPC title

• G10L15/1822

  Parsing for meaning understanding · CPC title

• G10L15/1815

  Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning · CPC title