Online Acoustic Simulations for Education and Listener Evaluations

Research Day Poster April 22, 2008
By: Travis Doll and Ray Migneco
Advisor: Dr. Youngmoo Kim

Abstract


Online collaborative activities provide a powerful platform for the collection of psychoacoustic data on the perception of audio and music from a very large numbers of subjects. Furthermore, these activities can be designed to simultaneously educate users about aspects of music information and acoustics, particularly for younger students in grades K-12. We have created prototype interactive activities illustrating aspects of two different sound and acoustics concepts: musical instrument timbre and the cocktail party problem (sound source isolation within mixtures) that also provide a method of collecting perceptual data related to these problems with a range of parameter variation that is difficult to achieve for large subject populations using traditional psychoacoustic evaluation. We present preliminary data from a pilot study where middle school students were engaged with the two activities to demonstrate the potential benefits as an education and data collection platform.


Game Structure


Both the Timbre Game and Cocktail Party game employ a client-server architecture that allows distributed online game play, so that players of both games may be widely separated. To maintain widespread accessibility, the activities require only internet access through a web browser and run independently of external applications.


Application Deployment


The MET-lab is currently involved with several K-12 outreach programs with suitable target audiences for incorporating our tools.
These outreach programs include:

  • NSF GK-12 Fellowship (Girard Academic Music Program)
  • Summer Music Technology (SMT)
  • NSF Discovery K-12 (Philadelphia Creative and Performing Arts H.S.)

Eventually, we plan to make our applications available through the internet so that we may collect data from a more diverse subject population.


Educational Objectives


The Cocktail Party Game will help students understand:

  • The relationship between the student’s distance from the speaker and reverberation intensity
  • How source locations and acoustic spaces affect the intelligibility of speech at different listening positions

The Timbre Game will help students understand:

  • Concepts of attack, decay, sustain and release by manipulating a time-varying amplitude envelope
  • How the weights of the underlying sinusoid components affect perceived sound
  • Time and spectral properties that characterize instruments and families of instruments


Cocktail Party Simulation


The Cocktail Party problem is characterized by the ability of humans to focus on one voice within a mixture of voices in a crowded room. This game explores this concept from two different approaches using the game interfaces.


Figure 1: The interface for the room creation component of the Cocktail Party Game.


Room Creation:

  • Provides a visual representation of room configuration
  • Illustrates the effects of reverberation and interfering sounds

Listening Room:

  • Evaluates the ability of a listener to detect a known person's voice within a mixture of voice’s in a room

Timbre Game


The Timbre Game illustrates musical timbre and collects evaluation data on timbre perception. The game features two user interfaces:


Figure 2: The interface for the sound creation component of the Timbre Game.


Timbre Modifier

  • Analyzes timbre of real instruments
  • Players modify the amplitude and spectral envelopes

Timbre Listener

  • Players listen to sounds modified by other players
  • Evaluates players’ tolerance to timbre modification

Results


Both games were tested at GAMP with 56 8th grade students under the following conditions:

  • Students worked individually using headphones
  • Given 20 minute sessions with each game component
  • Over 350 sounds were created for each game
  • Over 800 listening trials obtained from each game
  • Generally better detection accuracy with higher SNR
  • Students preferred listening over creation activities

Figure 3: Players’ accuracy for correctly detecting the presence of the speaker of interest versus Signal to Interference plus Noise Ratio.



Figure 4:Players’ accuracy for correctly identifying instruments and families with varying Signal to Noise Ratios.