EPSRC GR/S00859/01

Project
title

   

Intelligent and perceptual-based techniques for automated design and synthesis of musical instrument sounds

Time

 

Start 1st April 2003
Ended: 31 March 2006

Principal Investigator

Post-Doctorate
Research Fellow

 

Professor Emmanuel C. Ifeachor

Dr Brahim Hamadicharef

Emails

 

eifeachor@plymouth.ac.uk 
bhamadicharef@plymouth.ac.uk 

 

Abstract

Digital synthesis of musical instrument sounds is a key aspect of the rapidly growing field of computer music technology. It allows the creation of realistic sounds and continuous 'animation' beyond the physical boundaries of the real instrument. This is vital as it can, for example, inform the design and use of next generation musical instruments and aids the preservation of historical instruments. However, professional synthesis of instrument sounds is time consuming and requires a great deal of experience. This is expensive and limits productivity in the manufacture of digital synthesis systems, for example, in what is an important export market. The aim is to extend the underlying techniques of our new intelligent audio design system to provide a frame work for automated modelling and synthesis of acoustic instruments sounds to reduce work load. The outcome of the project include (i) a new model, based on fuzzy logic, that encodes existing skills base and audio expertise and uses these to automate the processing of sound features, (ii) a novel method that uses a perceptual-based measure to optimise sound synthesis parameters; and (iii) a new approach that provides an objective measure of perceptual quality of synthesised sound, based on the new ITU standard for wide band audio quality. Pipe organ will be used as a vehicle for the investigation because of its importance to the collaborating companies and in sound synthesis. Knowledge gained from synthesis of such a complex instrument will be invaluable in the synthesis of other instruments.

Funding

Starts: 1 April 2003 / Ends: 31 March 2006
Value: Ł140,218
Scheme: Standard Research
EPSRC Research Topic Classifications: Vision, Hearing and Other Senses - Applications in ICT

Industrial collaborators

Publications

Intelligent and Perceptual-based Approach to Musical Instruments Sound Design
Brahim Hamadicharef and Emmanuel C. Ifeachor
Submitted to the IEEE Transactions on Audio, Speech, and Language Processing (Manuscript No. T-ASL-01192-2006)
 
Perceptual Modeling of Piano Tones
Brahim Hamadicharef and Emmanuel C. Ifeachor
119th AES Convention, October 7-10, 2005 - New York, NY, USA, Preprint 6525
 
Objective Prediction of Sound Synthesis Quality
Hamadicharef B. and Ifeachor E. C.
Proceedings of the 115th AES Convention, Oct. 10-13, 2003 New York, USA, Preprint 5958
 
An Intelligent System Approach to Sound Synthesis Parameter Optimisation
Hamadicharef, B. and Ifeachor, E. C.
111th AES Convention, Nov-30 Dec 3, 2001, New York, USA, Preprint 5484
  

Developed Research Tools

Sound examples

Training and Experience (Project supervision)

  • Adrien Martin (2005) Bachelor of Science
    B.Sc. Thesis - Real-time DSP-based Audio Signal Processing
     
  • Nicolai Heilemann (2005) Master Diploma
    M.Sc. Thesis - Audio Quality Prediction for Digital Audio Broadcast
     
  • Dineshen Chuckravanen (2004) M.Sc.
    M.Sc. Thesis - Independent Sound Component Analysis
     
  • Simon Makadi Bawa (2004) M.Sc.
    M.Sc. Thesis - Fuzzy Audio Signal Processing
     
  • Michael Hess (2004) 2nd practical training term
    Project Report - Investigations into Video Quality Assessment of Real-Time Video Streaming
     
  • Charles Marzin (2003) M.Sc.
    M.Sc. Thesis - Prediction of Audio Quality for Personal Audio Devices
     
  • Yifan Zhang (2003) M.Res.
    M.Res. Thesis - Automated Modeling and Synthesis of Piano Sounds
     
  • Jeremie Grizard (2003) M.Sc.
    Thesis - Sound Synthesis using Independent Component Analysis
     
  • Sascha Tenkleve (2002-2003) Diploma Ing.
    Diploma Thesis - Perceptual Modeling of the Hammond Organ

EPSRC Web site

2006 - SPMC / SoCCE / UoP