Method and apparatus for processing languages

• US 4,468,756 A
• Filed: 05/20/1982
• Issued: 08/28/1984
• Est. Priority Date: 05/20/1982
• Status: Expired due to Fees

• Current Assignee: YIM FAT CHAN Parkview Convalescent Center Room
• Original Assignee: CHAN Computer Corporation
• Inventors: Chan, Yim F.
• Primary Examiner(s): Zache, Raulfe B.
• Application Number:
  US06/380,244
• Time in Patent Office: 831 Days
• Field of Search: 364/200 MS File, 364/900 MS File, 364/300
• US Class Current: 704/4
• CPC Class Codes: G06F 40/53 Processing of non-Latin text kana-to-kanji conversion G06F40/129; vowelisation G06F40/232
• CPC Class Codes: G10L 25/15 the extracted parameters being formant information

Abstract

A method for processing a spoken language of substantially monosyllabic words corresponding to individual, transcribable character codes of complex configuration, comprising the steps of: compiling a first, machine-readable library file of encoded signals for each of the words as spoken; compiling a second, machine-readable library file of those of the character codes uniquely corresponding to one of the spoken words and homonym set identifying codes for groups of those of the character codes corresponding to the same spoken word, for each of the encoded signals in the first library file; compiling a third, machine-readable library file of sets of homonyms for each of the homonym set identifying codes in the second library file; sequentially converting spoken words into the encoded signals, utilizing the first library file; sequentially displaying the information in the second library file corresponding to the encoded signal corresponding to the sequentially converted spoken word; and, if the displayed information is one of the homonym set identifying codes, displaying the set of homonyms from the third library file corresponding to the displayed homonym set identifying code, whereby the correct transcribable character code for each spoken word may be easily and quickly identified.

An apparatus for processing a spoken language of substantially monosyllabic words corresponding to individual, transcribable character codes of complex configuration, comprising: a first, machine-readable library file for storing and encoded signal for each spoken word; a second, machine-readable library file for storing those of the character codes uniquely corresponding to one of the spoken words and homonym set identifying codes for groups of those of the character codes corresponding to the same spoken word, for each of the encoded signals in the first library file; a third, machine-readable library file for storing sets of homonyms for each of the homonym set identifying codes in the second library file; a converter for sequentially converting each of the spoken words into signals for accessing the encoded signals in the first library file; and, a display for the information from the second library file corresponding to the accessed encoded signal, and for further displaying set of homonyms, from the third library file, whereby the correct transcribable character code for each spoken word may be easily and quickly identified.

Claims

1. A method for processing a spokenk language of substantially monosyllabic words corresponding to individual, transcribable character codes of complex configuration, comprising the steps of:

compiling a first, machine-readable library file of encoded signals for each of the words as spoken;

compiling a second, machine-readable library file or those of the character codes uniquely corresponding to one of the spoken words and homonym set identifying codes for groups of those of the character codes corresponding to the same spoken word, for each of the encoded signals in the first library file;

compiling a third, machine-readable library of sets of homonyms for each of the homonym set identifying codes in the second library file;

sequentially converting spoken words into the encoded signals, utilizing the first library file;

sequentially displaying the information in the second library file corresponding to the encoded signal corresponding to the sequentially converted spoken word; and

, if the displayed information is one of the homonym set identifying codes,

displaying the set of homonyms from the third library file corresponding to the displayed homonym set identifying code, whereby the correct transcribable character code for each spoken word may be easily and quickly identified.

2. The method of claim 1, further comprising the step of generating a command signal, corresponding to the correctly identified transcribable character code, for further processing.

3. The method of claim 2, further comprising the step of automatically and sequentially transcribing the correctly identified character codes in response to the command control signals.

4. The method of claims 1, 2 or 3, wherein the spoken words are sequentially converted into the encoded signals by the steps of:

acoustically discriminating between the spoken words by measuring frequencies and relative intensities of components of the sounds as a function of time;

generating analog signals corresponding to the measured frequencies and intensities; and

,

converting the analog signals into digital codes forming the encoded signals.

5. The method of claim 1, further comprising the step of:

compiling a fourth, machine-readable library file of a second set of transcribable character codes of a second language, corresponding to a translation for each of the character codes in the second library file, whereby the spoken language may be easily translated into the second language.

6. The method of claim 5, further comprising the step of:

sequentially and automatically transcribing the translated words.

7. An apparatus for processing a spoken language of substantially monosyllabic words corresponding to individual, transcribable character codes of complex configuration, comprising:

a first, machine-readable library file for storing an encoded signal for each spoken word;

a second, machine-readable library file for storing those of the character codes uniquely corresponding to one of the spoken words and homonym set identifying codes for groups of those of the character codes corresponding to the same spoken word, for each of the encoded signals in the first library file;

a third, machine-readable library file for storing sets of homonyms for each of the homonym set identifying codes in the second library file;

means for sequentially converting each of the spoken words into signals for accessing the encoded signals in the first library file; and

,

means for displaying information from the second library file corresponding to the accessed encoded signal, and for further displaying one of the sets of homonyms, from the third library file, whereby the correct transcribable character code for each spoken word may be easily and quickly identified.

8. The apparatus of claim 7, further comprising a fourth, machine-readable library file for storing a second set of transcribable character codes of a second language corresponding to a translation for each character code, whereby the spoken words may be easily and quickly translated into a second language.

9. The apparatus of claims 7 or 8, further comprising means for generating a command signal, corresponding to the correctly identified transcribable character code, for further processing.

10. The apparatus of claims 7 or 8, further comprising means for automatically and sequentially transcribing the correctly identified character codes.

11. A method for processing a spoken language of words corresponding to individual, transcribable character codes of complex configuration, comprising the steps of:

compiling a machine-readable reference library, having a first file of encoded signals for each of the words as spoken, a second file having an entry for each of the encoded signals in the first file, of;

the character codes uniquely corresponding to one of the spoken words; and

, homonym set identifying codes for any other of the character codes corresponding to the same spoken word, and a third file of sets of homonyms for each of the homonym set identifying codes in the second file;

sequentially processing spoken words into machine-readable codes and comparing the codes with the encoded signals in the first file;

sequentially displaying the information in the second file corresponding to the encoded signal closest to the machine-readable code; and

, when the displayed information is one of the homonym set identifying codes,

displaying the set of homonyms from the third file corresponding to the displayed homonym set identifying code, whereby the correct transcribable character code for each spoken word may be easily and quickly identified.

12. The method of claim 11, further comprising the step of generating a command signal, corresponding to the correctly identified transcribable character code, for further processing.

13. The method of claim 12, further comprising the step of automatically and sequentially transcribing the correctly identified character codes in response to the command control signals, whereby the spoken words are easily converted into text.

14. The method of claims 11, 12 or 13, wherein the spoken words are sequentially processed into the encoded signals by the steps of:

acoustically discriminating between the spoken words by taking measurements of relative intensities of a range of frequency components of the sounds at a plurality of sampling times; and

,

converting the measurements into digital codes.

15. The method of claim 11, further comprising the step of:

compiling a fourth file in the machine-readable library of a second set of transcribable character codes of a second language, corresponding to a translation for each of the character codes in the second file, whereby the spoken language may be easily translated into the second language.

16. The method of claim 15, further comprising the step of:

sequentially and automatically transcribing the translated words.

17. An apparatus for processing a spoken language of words corresponding to individual, transcribable character codes of complex configuration, comprising:

a machine-readable reference library having a first file for storing an encoded signal for each spoken word;

a second file for storing an entry for each of the encoded signals in the first file, of;

the character codes uniquely corresponding to one of the spoken words; and

, homonym set identifying codes for any other of the character codes corresponding to the same spoken word, and a third file for storing sets of homonyms for each of the homonym set identifying codes in the second file;

means for sequentially processing each of the spoken words into machine-readable codes for comparison with the encoded signals in the first library file; and

,

means for displaying information from the second file corresponding to the accessed encoded signal closest to the machine-readable code, and for further displaying one of the sets of homonyms from the third file when necessary, whereby the correct transcribable character code for each spoken word may be easily and quickly identified.

18. The apparatus of claim 17, wherein the reference library has a fourth file for storing a second set of transcribable character codes corresponding to characters in a second language, the characters in the second language being translations of the data in the second file and the homonyms in the third file.

19. The apparatus of claims 17 or 18, further comprising means for generating a command signal, corresponding to the correctly identified transcribable character code, for further processing.

20. The apparatus of claims 17 or 18, further comprising means for automatically and sequentially transcribing the correctly identified character codes, whereby the spoken words may be easily converted into text.