Method and means for creating prosody in speech regeneration for laryngectomees

• US 6,795,807 B1
• Filed: 08/17/2000
• Issued: 09/21/2004
• Est. Priority Date: 08/17/1999
• Status: Active Grant

• Current Assignee: David R. Baraff
• Original Assignee: David R. Baraff
• Inventors: Baraff, David R.
• Primary Examiner(s): Dorvil, Richemond
• Assistant Examiner(s): Nolan, Daniel A
• Application Number:
  US09/641,157
• Time in Patent Office: 1,496 Days
• Field of Search: 704/205-210, 704/258-269, 704/270-271, 704/248, 704/234, 704/214
• US Class Current: 704/271
• CPC Class Codes: G10L 13/04 Details of speech synthesis systems, e.g. synthesiser structure or memory management
• CPC Class Codes: G10L 2021/0575 Aids for the handicapped in speaking
• CPC Class Codes: G10L 21/0208 Noise filtering

Abstract

A device and a method to be used by laryngeally impaired people to improve the naturalness of their speech. An artificial sound creating mechanism which forms a simulated glottal pulse in the vocal tract is utilized. An artificial glottal pulse is compared with the natural spectrum and an inverse filter is generated to provide an output signal which would better reproduce natural sound. A digital signal processor introduces a variation of pitch based on an algorithm developed for this purpose; i.e. creating prosody. The algorithm uses primarily the relative amplitude of the speech signal and the rise and fall rates of the amplitude as a basis for setting the frequency of the speech. The invention also clarifies speech of laryngectomees by sensing the presence of consonants in the speech and appropriately amplifying them with respect to the vowel sounds.

Claims

1. A method of creating or reproducing prosody in speech using a Linear Predictive Coding algorithm, comprising the steps of:

dividing speech to be processed into components of silent, consonant and vowel;

processing said silent component to determine a threshold level to alter said component to consonant sound or to maintain silent sound;

wherein further said consonant component is selected from a threshold value to determine whether said consonant component exceeds a threshold to be modified to a vowel, or selected for additional threshold measurement to change said consonant component from a consonant to a vowel;

wherein further said vowel component is measured against a threshold level set to determine whether said vowel component is changed from a vowel to a consonant.

2. A means for creating or reproducing prosody in speech comprising:

an analog to digital converting means to convert analog human speech to a digital equivalent;

a digital signal processor to process said digital equivalent signal;

an electronic memory means to store an instruction set to operate said digital signal processing means;

means to process said digital signal processor output to convert said output to a reconditioned analog voice signal; and

an instruction set stored in said electronic memory means to control said processing by said digital signal processor to alter the reconditioned analog voice signal in accordance with the intended sound of the speech being processed;

wherein further said digital signal processing means selects the input to said digital signal processing means to alternate and select between silent, consonant and vowel components of the-inputted human speech being processed;

wherein further, the silence component is capable of being further divided into silence or a consonant sound;

wherein the consonant component is capable of being further divided into silence or, upon reaching another pre-set threshold level, into a vowel sounds or a consonant sound;

wherein the vowel component is processed to be further divided into a consonant sounds or a vowel sound.