Electronic apparatus from a host language

• US 4,507,750 A
• Filed: 05/13/1982
• Issued: 03/26/1985
• Est. Priority Date: 05/13/1982
• Status: Expired due to Term

• Current Assignee: Texas Instruments, Inc.
• Original Assignee: Texas Instruments, Inc.
• Inventors: Lin, Kun-Shan, Frantz, Gene A.
• Primary Examiner(s): Smith, Jerry
• Assistant Examiner(s): Fleming, Michael R.
• Application Number:
  US06/377,734
• Time in Patent Office: 1,048 Days
• Field of Search: 364/419, 364/900, 364/513, 364/513.5, 381/41, 381/49, 381/42, 381/50, 381/43, 381/51, 381/44, 381/52, 381/45, 381/53, 381/46, 381/47, 381/48
• US Class Current: 704/277
• CPC Class Codes: G06F 40/40 Processing or translation of natural language natural language analysis G06F40/20; semantic analysis G06F40/30

Abstract

An electronic apparatus for translation from a host language to a non-host language in which the individual word is evaluated as to its contextual meaning. The sequence of words, typically a sentence, within the host language, which is communicated to the electronic apparatus is translated, through a recognition device into a series of recognized words. These recognized words are further refined through analysis of their contextual meaning within the sequence (sentence) so as to differentiate between words of similar pronunciation and between homonyms. The present invention permits the direct entry, from voice, to a translator to a foreign language or alternatively to control language for use with an electronic or electromechanical apparatus.

Claims

1. An electronic apparatus comprising:

memory means having digital data stored therein including a first plurality of digital speech data templates respectively representative of synthesized speech in a host language as grouped in speech segments of the host language, and a second plurality of digital data templates representative of a non-host language and having corresponding counterparts in said first plurality of digital speech data templates representative of synthesized speech in the host language;

operator input means including at least speech input means for receiving at least one word as spoken speech from an operator, said speech input means converting the operator-generated speech into an analog speech signal representative of a sequence of speech segments spoken by the operator;

speech analysis means operably connected to said operator input means for analyzing said analog speech signal and generating a sequence of speech data sets representative thereof;

control means operably connected to said speech analysis means and to said memory means and including speech recognition means for matching each respective speech data set from said speech analysis means with a particular digital speech data template of said memory means in response to the respective speech data set and the remaining speech data sets within said sequence of speech data sets and providing as a control signal output the digital data template of said second plurality of digital data templates representative of the non-host language; and

output means coupled to said control means for receiving the control signal output therefrom.

2. An electronic apparatus as set forth in claim 1, wherein said memory means includes allophone library means in which digital signals representative of allophone-defining speech parameters identifying the respective allophone subset variants of each of the recognized phonemes in the host language as modified by the speech environment in which the particular phoneme occurs are stored, said allophone library means being responsive to said speech recognition means of said control means to provide digital signals representative of the particular allophone-defining speech parameters corresponding to a respective speech data set from said speech analysis means.

3. An electronic apparatus as set forth in claim 2, wherein said allophone library means includes stored digital signals representative of allophone-defining speech parameters identifying the respective allophone subset variants of each of the recognized phonemes in at least one non-host language in addition to the host language ad modified by the speech environment in which the particular phoneme occurs.

4. An electronic apparatus as set forth in claim 1, wherein said memory means further includes a memory portion containing digital signals representative of grammatical rules and respectively determinative of the appropriate location of a word in the host language in a word sequence in the host language, said memory portion operating in conjunction with said speech recognition means of said control means in matching each respective speech data set from said speech analysis means with a particular digital speech data template of said memory means so as to provide an appropriate sequence of speech data sets definitive of a particular word order in a given word sequence.

5. An electronic apparatus as set forth in claim 1, wherein said output means comprises speech synthesizer means coupled to said control means for receiving the control signal output therefrom in the non-host language and providing analog signals representative of synthesized speech in the non-host language;

and

audio output means operably connected to the output of said speech synthesizer means for receiving said analog signals representative of synthesized speech in the non-host language to produce audible synthesized speech in the non-host language.

6. An electronic apparatus as set forth in claim 5, wherein the non-host language is a foreign language such that the control signal output comprises digital data representative of the foreign language equivalent of the spoken speech input from the operator at said operator input means.

7. An electronic apparatus as set forth in claim 1, wherein said output means comprises an electro-mechanical apparatus, and said non-host language comprises a command as a control signal output to said electro-mechanical apparatus for regulating the operation thereof.

8. An electronic apparatus as set forth in claim 7, wherein said electro-mechanical apparatus is a typewriter.

9. An electronic apparatus as set forth in claim 1, further including means operably associated with the output of said control means for converting the character of said control signal output therefrom into a programming language instruction set for reception by said output means.

10. An electronic apparatus as set forth in claim 1, wherein said memory means includes dictionary means in which digital speech data is stored comprising at least first and second tables of digital speech data representative of speech segments in a host language and in a non-host language;

and

said control means being responsive to the matching of each respective speech data set from said speech analysis means with a particular digital speech data template in said memory means corresponding to digital speech data included in said first table of said dictionary means for providing as a control signal output the digital speech data included in said second table of said dictionary means and comprising the non-host language equivalent to the host language.

11. An electronic language translator comprising:

memory means having digital data stored therein including a first plurality of digital speech data templates respectively representative of synthesized speech in individual words of a host language, and a second plurality of digital data templates respectively representative of synthesized speech in individual words of a non-host language and having corresponding counterparts in said first plurality of digital speech data templates representative of synthesized speech in individual words of the host language;

said memory means further including a memory portion having digital data stored therein representative of a plurality of grammatical rules relating to the syntactical significance of word sequences in the host and non-host languages and being determinative of the location of a given word in a word sequence occurring in the host and non-host languages;

operator input means including at least speech input means for receiving at least one word as spoken speech in the host language from an operator, said speech input means converting the operator-generated speech into an analog speech signal representative of a sequence of speech segments in the host language as spoken by the operator;

speech analysis means operably connected to said operator input means for analyzing said analog speech signal and generating a sequence of speech data sets in the host language representative thereof;

control means operably connected to said speech analysis means and to said memory means and including speech recognition means for matching each respective speech data set from said speech analysis means with a particular digital speech data template of said memory means in the host language in response to the respective speech data set and the remaining speech data sets within said sequence of speech data sets as determined by said memory portion in which said digital data representative of grammatical rules are stored;

said control means being responsive to a matched sequence of speech data sets in said host language as determined by said speech recognition means for accessing digital data templates from said memory means representative of the non-host language equivalent to the matched first plurality of digital speech data templates in the host language and providing the digital data templates representative of the non-host language as a control signal output therefrom;

the non-host language being a foreign language such that the control signal output comprises digital data representative of the foreign language equivalent of the spoken speech input from the operator at said operator input means; and

output means coupled to said control means for receiving the control signal output therefrom and communicating the control signal output to the operator as the foreign language equivalent of the spoken speech input.

12. An electronic language translator as set forth in claim 11, wherein said output means comprises speech synthesizer means coupled to said control means for receiving the control signal output therefrom as digital data representative of the foreign language equivalent of the spoken speech input from the operator at said operator input means and providing analog signals representative of synthesized speech in the foreign language;

and

audio output means operably connected to the output of said speech synthesizer means for receiving said analog signals representative of synthesized speech in the foreign language to produce audible synthesized speech in the foreign language.