Orchestration and Timbral Analysis

Music 255 - Winter 2002
cd usr/cc

Monday 2:15-3:45 - CCRMA Ballroom

Instructor: Jonathan Berger

Teaching assistants: Randal Leistikow, Unjung Nam

lecture notes


Course Goals

The goals of this course are to:

Required skills

It is assumed that you have music theory skills at the level of music 22 or beyond.

In addition, by the end of the course you will be expected to know the mechanics of

sound production and control of each orchestral instrument. Proficiency in notation,

particularly as relavent to orchestration, is also expected.


Course Syllabus

timbre defined

vagueries of common music notation of timbre

developing terminology for timbre description

time domain

frequency domain

issues of time and frequency domain resolution-

instrument constraints

performer limitations

economics of the orchestra


constraints of the performing space

analysis of Mahler's reorchestrations

scalars, vectors and matrices



audio I/O

An intuitive introduction to sinusoidal sectral modeling
Fourier representations of discrete musical events - DFT

Refinements of the FFT - windowing, filtering



plotting methods

the spectrogram and sonogram


Analyzing the human voice - Cepstral analysis - MFCC

Replicating the human voice

vowels and consonants

synthetized voices

orchestrated voices - project goals

percussion instruments and consonants

diracs and glissandi

Orchestra readings: March 6th


The final project will involve synthesizing the techniques and skills acquired in the class with an orchestration excercise that will be read by the Stanford Symphony on March 6 2002.


The excercise will involve making the orchestra 'speak' a text by applying principles of vowel production and spectral analysis to find close approximations between orchestral sounds and specific consonants and vowels.