Automatic multi-language phonetic transcribing system

• US 20030191643A1
• Filed: 04/03/2002
• Published: 10/09/2003
• Est. Priority Date: 04/03/2002
• Status: Active Grant

• Current Assignee: the united states of america as represented by the secretary of the navy
• Original Assignee: the united states of america as represented by the secretary of the navy
• Inventors: Belenger, Robert V., Lopriore, Gennaro R.
• Granted Patent: US 7,143,033 B2
• Time in Patent Office: Days
• US Class Current: 704/254
• CPC Class Codes: G10L 15/005 Language recognition
• CPC Class Codes: G10L 15/187 Phonemic context, e.g. pronunciation rules, phonotactical constraints or phoneme n-grams
• CPC Class Codes: G10L 2015/025 Phonemes, fenemes or fenones being the recognition units

Abstract

A multi-language phonetic transcribing system and method are provided for automatically transcribing speech into a phonetic equivalent. The phonetic transcribing system need only be able to recognize the limited number of phonemes of a particular language, e.g., the forty-two phonemes in the English language. Each language to be translated may be broken down into the phonetic elements and stored in a phonetic library for that language. The transcribing system detects speech, converts the speech to an electrical signal, and analyzes the frequency, amplitude, and timing characteristics of the speech to produce incoming phoneme information. The incoming phoneme information is compared to the phonetic elements in the active library. A correlator determines the degree of correlation between the incoming phoneme information and the stored information for each phoneme in the library. For each match, based on the degree of correlation, a respective phoneme is preferably stored and/or printed.

Claims

1. A transcriber for transcribing speech, said transcriber comprising:

a sound transducer for converting said speech to an electrical signal;

a speech analyzer operable for analyzing said electrical signal and producing a digital signal containing incoming phoneme information;

at least one phonetic library stored in a digital memory, said at least one phonetic library storing a plurality of phonemes; and

a phoneme sound correlator operable for comparing said digital signal containing incoming phoneme information to said plurality of phonemes stored in said phonetic library.

2. The transcriber of claim 1 further comprising a match detector for indicating a match between said incoming phoneme information and said plurality of phonemes stored in said phonetic library.

3. The transcriber of claim 2 wherein said match detector is operable for determining said match between said incoming phoneme information and said plurality of phonemes stored in said phonetic library by comparison with a predetermined level of correlation.

4. The transcriber of claim 3 further comprising a phoneme buffer for storing detected phonemes in response to an indication of said match between said incoming phonemes and said plurality of phonemes stored in said phonetic library.

5. The transcriber of claim 4 further comprising a printer for printing phonetic equivalents of said detected phonemes.

6. The transcriber of claim 5 further comprising a phoneme encoder for converting said detected phonemes to a format compatible with said printer.

7. The transcriber of claim 1 wherein said speech analyzer is operable for measuring an amplitude and frequency content related to said electrical signal to produce said incoming phoneme information.

8. The transcriber of claim 7 wherein said speech analyzer is operable for associating a timing sequence with said amplitude and frequency content to produce said incoming phoneme information.

9. The transcriber of claim 8 wherein said speech analyzer is operable for digitizing said amplitude and frequency content and said timing sequence to produce said digital signal containing incoming phoneme information.

10. A method for transcribing speech, comprising:

storing a plurality of phonemes in a digital format in one or more phoneme libraries;

converting said speech to an electrical signal;

processing said electrical signal to produce a digital signal;

comparing said digital signals to said plurality of phonemes stored in said one or more phoneme libraries for determining a plurality of matches therebetween; and

storing a plurality of phonemes in response to said determining of said plurality of matches.

11. The method of claim 10 further comprising:

providing a plurality of phoneme libraries; and

selecting a phoneme library corresponding to a language of said speech.

12. The method of claim 10 wherein said step of processing further comprises detecting frequency, amplitude, and timing information from said electrical signal to produce said digital signal.

13. The method of claim 12 wherein said step of comparing further comprises comparing said frequency, amplitude, and timing information in said digital signal to frequency, amplitude, and timing information related to said plurality of phonemes stored in said one or more phoneme libraries for determining said plurality of matches.

14. The method of claim 13 further comprising determining a correlation factor in response to said step of comparing.

15. The method of claim 14 comparing said correlation factor with a predetermined correlation factor for determining said plurality of said matches.

16. A transcriber for transcribing speech, said transcriber comprising:

a sound transducer for converting said speech to an electrical signal;

a speech analyzer operable for producing a digital signal containing incoming phoneme information related to an amplitude, frequency, and timing of said speech;

at least one phonetic library stored in a digital memory, said at least one phonetic library storing a plurality of phonemes; and

a match detector for determining a match between said incoming phoneme information and said plurality of phonemes stored in said phonetic library.

17. The transcriber of claim 16 further comprising a phoneme sound correlator for determining a degree of correlation between said incoming phoneme information and said plurality of phonemes stored in said phonetic library.

18. The transcriber of claim 16 further comprising a plurality of phoneme libraries, and a switch for selecting a particular phoneme library corresponding to a language of said speech.

19. The transcriber of claim 16 further comprising a phoneme buffer for storing detected phonetic phonemes in response to an indication of said match between said incoming phoneme information and said plurality of phonemes stored in said phonetic library.

20. The transcriber of claim 16 further comprising a printer for printing phonetic equivalents of said detected phonemes.