Pitch frequency generation system in a speech synthesis system

• US 4,907,279 A
• Filed: 07/11/1988
• Issued: 03/06/1990
• Est. Priority Date: 07/31/1987
• Status: Expired due to Fees

• Current Assignee: Kokusai Denshi Denwa Company Limited (KDDI Corporation)
• Original Assignee: Kokusai Denshi Denwa Company Limited (KDDI Corporation)
• Inventors: Yamamoto, Seiichi, Shimizu, Toru, Higuchi, Norio
• Primary Examiner(s): Harkcom, Gary V.
• Assistant Examiner(s): NOT, DEFINED
• Application Number:
  US07/217,520
• Time in Patent Office: 603 Days
• Field of Search: 381/51-53, 381/36-40, 364/513.5
• US Class Current: 704/260
• CPC Class Codes: G10L 13/08 Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination

Abstract

A speech synthesis system comprises an input terminal for accepting text code, accent code, and phrase code. The speech synthesis system further comprises a converter for converting the text code to speech parameters for speech synthesis; an accent commend generator coupled to an output of the converter for providing a train of accent commands; a phrase command generator coupled to an output of the converter for providing a train of phrase commands; an accent command buffer for storing the accent commands; a phrase command buffer for storing the phrase commands; an accent component calculator operably coupled to the accent command buffer for providing contour of pitch frequency by accent component; a phrase component calculator operably coupled to the phrase command buffer for providing contour of pitch frequency by phrase component; an adder for providing a sum of output signals from the accent component calculator and the phrase component calculator; a device for providing fundamental frequency of voicing which is coupled to an output of the adder; a speech synthesizer coupled to an output of the device for providing the fundamental frequency and output of the converter; and an output terminal coupled to an output of the speech synthesizer for providing synthesized speech to an external circuit.

Claims

1. A speech synthesis system, comprising:

an input terminal means for accepting text code, said text code being at least a character train having spelling, accent code, and phrase code of each word;

generating means for receiving said text code from said input terminal means and for converting said text code to thereby generate speech parameters for speech synthesis;

an accent command generator means for receiving said accent code which is one of said speech parameters from said generating means, and for providing a train of accent commands, each accent command being defined by a start point time, an end point time, and an amplitude which define a step function;

a phrase command generator means for receiving said phrase code which is one of said speech parameters from said generating means, and for providing a train of phrase commands, each phrase command being defined by a time and an amplitude which define an impulse function;

an accent command buffer means for storing said accent commands;

a phrase command buffer means for storing said phrase commands;

an accent component calculator means coupled to said accent command buffer means for providing an output signal representing an accent component;

a phrase component calculator means coupled to said phrase command buffer means for providing an output signal representing a phrase component;

an adder means for providing a sum of said output signals from said accent component calculator means and said phrase component calculator means;

means for receiving said sum provided by said adder means, and for providing fundamental frequency of voicing;

a speech synthesizer means coupled to said means for providing said fundamental frequency and said generating means for providing synthesized speech; and

an output terminal means coupled to said speech synthesizer means for providing said synthesized speech to an external circuit,

wherein said accent component calculator means comprises at least one accent table which stores a response for a step function corresponding to at least one accent command from said accent command generator, wherein said accent component calculator means adds together step function responses stored in at least one accent table corresponding to at least one accent command received from said accent command buffer means and provides the result of the additional which is outputted as said output signal representing said accent component, and

wherein said phrase component calculator means comprises a single phrase table for storing impulse response for a unit amplitude, and a multiplier means for providing a product of an output of said table and said amplitude of each phrase command from said phrase command generator, said product being outputted as said output signal representing said phrase component.

2. A speech synthesis system, according to claim 1, wherein said accent component calculator means has a plurality of accent tables, wherein each accent table is for storing a response for a step function according to said amplitude of an accent command, and an adder means for providing a sum of the outputs of said accent tables, said sum being outputted as said output signal representing said accent component.

3. A speech synthesis system, according to claim 2, wherein the number of accent tables in said accent command calculator means is six.

4. A speech synthesis system, according to claim 1, wherein said phrase table in said phrase component calculator means has a first portion, and a second portion which follows said first portion, said first portion stores relations between time and amplitude of an impulse response, and said second portion stores relations between n and M_n, where n is an address for M_n and is an integer greater than zero, and M_n is an end point time in which an impulse response decreases from n to n-1 by one unit.

5. A speech synthesis system, according to claim 1, wherein said start point time, and an said end point time of each accent command are defined by a number of frames which have a predetermined time duration.

6. A speech synthesis system according to claim 5, wherein, the predetermined time duration of each frame is 5 msec.

7. A speech synthesis system according to claim 1, wherein said accent command buffer means is for storing said accent commands for 500 msec.

8. A speech synthesis system according to claim 1, wherein said accent commands stored in said accent command buffer means are deleted after a predetermined time elapses, and a phrase command stored in said phrase buffer is deleted after another predetermined time elapses.

9. A speech synthesis system according to claim 1, wherein a time constant of said step function is the same as a time constant of said impulse function.

10. A speech synthesis system according to claim 9, wherein said time constants are between 15 msec and 30 msec.