Sound characteristic analyzer with a voice characteristic classifying table, for analyzing the voices of unspecified persons

• US 5,774,850 A
• Filed: 04/26/1996
• Issued: 06/30/1998
• Est. Priority Date: 04/26/1995
• Status: Expired due to Term

• Current Assignee: FUJITSU LIMITED & ANIMO LIMITED
• Original Assignee: FUJITSU LIMITED & ANIMO LIMITED
• Inventors: Hattori, Ichiro, Suzuki, Akira
• Primary Examiner(s): MacDonald, Allen R.
• Assistant Examiner(s): Dorvil, Richemond
• Application Number:
  US08/638,295
• Time in Patent Office: 795 Days
• Field of Search: 395/2.55, 395/2.5, 395/2.79, 395/2.86, 395/2.59, 395/2.56, 395/2.58, 395/2.6, 704/250, 704/272, 704/246, 704/241, 704/270, 704/277, 704/271, 704/249, 704/251
• US Class Current: 704/250
• CPC Class Codes: G10L 17/26 Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices

Abstract

A voice characteristics analyzer analyzes the voices of unspecified persons through conventional Karaoke systems. A voice characteristics extraction device extracts spoken voice characteristics when a user speaks specific words. A voice characteristic classification table pre-stores analysis information derived from non-user voices and corresponding to voice characteristics of these specific words. An analysis information output device outputs results of the characteristics analysis that compares current user data with the pre-stored information.

Claims

1. A sound characteristic analyzer comprising:

requesting means for requesting a user to speak predetermined words by displaying the predetermined words on a display;

input means for inputting by a user of spoken sound of predetermined words;

extracting means for extracting voice characteristics of said spoken sound of predetermined words inputted by said input means;

a voice characteristics classifying table for prestoring analysis information corresponding to a plurality of voice characteristics of said predetermined words, said prestored analysis information being derived from spoken sounds of others than said user; and

an analysis information output means for outputting analysis information, based upon said voice characteristics classifying table and said voice characteristics extracted from said user spoken sound by said extracting means.

2. A sound characteristic analyzer as in claim 1, wherein said voice characteristics classifying table stores analysis information on health conditions corresponding to said voice characteristics.

3. A sound characteristic analyzer as in claim 1, wherein said voice characteristics classifying table stores analysis information on individual human traits corresponding to said voice characteristics.

4. The sound characteristics analyzer as in claim 1, wherein said input means is a Karaoke microphone connected to said voice characteristics analyzer.

5. The sound characteristics analyzer as in claim 1, wherein said predetermined words contain many vowel sounds.

6. The sound characteristics analyzer as in claim 1, wherein said voice characteristics are classified according to voice speaking duration of predetermined words.

7. The sound characteristics analyzes as in claim 1, wherein said voice characteristics are classified by pitch characteristics of specific words of said inputted predetermined words.

8. The sound characteristics analyzer as in claim 1, wherein said voice characteristics are classified by sonargram information, that includes the time distribution of said extracted voice characteristics by frequency bandwidth.

9. A sound characteristic analyzer comprising:

input means for inputting by users of spoken sounds of predetermined words;

extracting means for extracting voice characteristics of said user spoken sounds of the predetermined words inputted by said input means;

a suitableness degree classifying table for storing analysis information corresponding to a relationship between two human voice characteristics, and

an analysis information output means for outputting analysis information based upon said suitableness degree classifying table and said two human voice characteristics extracted by said extracting means from said user spoken sounds of predetermined words.