Jitka Trojankova (trojaj2@fel.cvut.cz) Ivo Jirele - - PowerPoint PPT Presentation

IBM - CVUT Student Research Projects Physiological measurements of non-speech input

Jitka Trojankova (trojaj2@fel.cvut.cz) Ivo Jirele (jireli1@fel.cvut.cz) Lukas Wroblewski (wrobll1@fel.cvut.cz)

IBM - CVUT Student Research Projects 2

Goal

• Quantify the ability of users to “aim” the

pitch of tone they produce

• Non-speech input as an addition to voice

recognition for faster response

– Realtime applications, games, etc.

• Existing prototypes but quantification

research has not been done yet

IBM - CVUT Student Research Projects 3

Research Questions

• Range Measurement

– What is typical range of pitch that users can produce and use?

• Tone Released Within Given Interval

– How many different pitches can be recognized?

• Speed of Re-intonation

– How quickly can users slide from one pitch to another?

IBM - CVUT Student Research Projects 4

Number of Pitches?

wanted signal produced signal problem time pitch

IBM - CVUT Student Research Projects 5

Conditions

• Different ages and genders of participants
• 25 participants, average age 30 years,

standard deviation 15 years

• 20 - 40 minutes per person
• Input through humming (“mmm...”)
• Recording for later analysis

IBM - CVUT Student Research Projects 6

Range Measurement

2 4 6 8 10 >10 2,5 5 7,5 10 12,5 15 17,5 20 22,5 25 27,5 30

roughness of range borders

range in halftones percent of people

<10 10—13 14—17 18—21 >21 10 20 30 40 50 60 70 80 90 100

comfortable range width

users' range in halftones percent of people

IBM - CVUT Student Research Projects 7

Tone Released within an Interval

• From range=12 and width=4, we get 3 recognizable pitches

1 2 4 8 10 20 30 40 50 60 70 80 90 100

success in targetting specific interval width

width of target interval in halftones percent of success

<5 5—10 10—15 >15 10 20 30 40 50 60 70 80 90 100

distance from interval when failed

distance in halftones percent of failed tasks

IBM - CVUT Student Research Projects 8

Speed of Re-intonation

8 10 14 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000

time needed to slide into interval 8 halftones wide

distance of target interval in halftones time [ms] 6 8 12 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000

time needed to slide into interval 4 halftones wide

distance of target interval in halftones time [ms]

IBM - CVUT Student Research Projects 9

Speed of Re-intonation

5 7 11 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400

time needed to slide into interval 2 halftones wide

distance of target interval in halftones time [ms] 4 - 5 6 - 7 10 - 11 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800

time needed to slide into interval 1 halftone wide

distance of target interval in halftones time [ms]

IBM - CVUT Student Research Projects 10

Research Answers

• Typical users' range they can produce and

use is an octave

• There are 3 pitches that can be safely

recognized

• Sliding from one pitch to another takes

about 1 to 2 seconds, depending on distance and width of target interval