System and method for speech synthesis employing improved formant composition

• US 5,325,462 A
• Filed: 08/03/1992
• Issued: 06/28/1994
• Est. Priority Date: 08/03/1992
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Farrett, Peter W.
• Primary Examiner(s): MacDonald, Allen R.
• Assistant Examiner(s): KIM, RICHARD
• Application Number:
  US07/923,635
• Time in Patent Office: 694 Days
• Field of Search: 381/51, 381/52, 395/2.67, 395/2.77
• US Class Current: 704/258
• CPC Class Codes: G10L 13/033 Voice editing, e.g. manipulating the voice of the synthesiser
• CPC Class Codes: G10L 21/0364 for improving intelligibility
• CPC Class Codes: G10L 25/15 the extracted parameters being formant information

Abstract

A method, system and process to improve the formant composition in a speech synthesis system so that the formants are more intelligible. The system employs a process in the memory of a processor to change the starting and ending frequency of phonemes from the frequency of the independent phonemes. The process examines preceding and succeeding ending phoneme frequency values to detect similar phoneme frequency values. If a dissimilar value is detected, then the invention provides for exchange of the formants to render the resulting speech more intelligible.

Claims

1. A speech synthesis apparatus, comprising:

(a) memory means for receiving a plurality of data blocks each representing unit of speech information;

(b) means for identifying and parsing at least a first formant in a first one of said data blocks and a second formant in a second one of said data blocks;

(c) means for comparing the first formant and the second formant;

(d) means for replacing the first formant in a portion of said first data block by said second formant and replacing the second formant in a portion of said second data block by said first formant if the first formant and the second formant do not match; and

(e) means for synthesizing the plurality of data blocks into audio signals.

2. An apparatus as recited in claim 1, including a digital signal processor for processing the unit of speech information.

3. An apparatus as recited in claim 1, including analog to digital conversion means for receiving audio signals and converting them to information that a computer can process.

4. An apparatus as recited in claim 1, including digital to analog conversion means for receiving audio signals that a computer can process and converting it to analog audio signals.

5. An apparatus as recited in claim 1, including means for storing the unit of speech information.

6. A method for speech synthesis, comprising the steps of:

(a) receiving a plurality of data blocks each representing unit of speech information;

(b) identifying and parsing at least a first formant in a first one of said data blocks and a second formant in a second one of said data blocks;

(c) comparing the first formant and the second formant;

(d) replacing the first formant in a portion of said first data block by said second formant and the second formant in a portion of the second data block by said first formant if the first formant and the second formant do not match; and

(e) synthesizing the plurality of data blocks into audio signals.

7. A method as recited in claim 6, including the step of processing the unit of speech information with a digital signal processor.

8. A method as recited in claim 6, including the step of converting analog signals to digital information that a computer can process.

9. A method as recited in claim 6, including the step of receiving audio information that a computer can process and converting it to analog audio signals.

10. A method as recited in claim 6, including the step of storing the audio information.