A musician claims that two instruments playing the same pitch sound different only because of their attack transients — the sustained portion is identical. What does Fourier analysis of the sustained tone reveal?
AThe musician is correct — sustained tones at the same pitch are identical in spectral content
BThe instruments share the same fundamental frequency but have different harmonic spectra — the amplitudes of their overtones differ, creating distinct waveforms
CThe instruments produce inharmonic partials rather than true harmonics, so Fourier analysis does not apply to sustained tones
DThe sustained tone of one instrument contains more harmonics in total, while the other has fewer
Timbre is encoded in the harmonic spectrum: which overtones are present and at what relative amplitudes. A clarinet, violin, and oboe playing A440 all have the same fundamental (440 Hz), but their waveforms differ in the energy distribution across the 2nd, 3rd, 4th harmonics and beyond. Fourier analysis decomposes each sustained tone into this series, revealing the amplitude pattern that distinguishes them perceptually — not just in the attack. Attack transients matter too, but the spectral envelope of the sustained portion is a primary contributor to timbre identity.
Question 2 Multiple Choice
A researcher applies a single Fourier transform to an entire 3-minute symphony recording. Which is the fundamental limitation of this approach?
AThe Fourier transform can only decompose signals up to a limited maximum frequency
BAll temporal information is lost — the analysis cannot show how the spectrum changes from moment to moment throughout the recording
CThe Fourier transform requires the signal to be exactly periodic, and a symphony is not
DThe technique is computationally too expensive for signals longer than a few seconds
A single Fourier transform applied to the whole recording collapses all temporal structure into one averaged spectrum. A chord played at the beginning and a different chord played later both contribute to the same frequency bins simultaneously, making it impossible to track how timbre and harmony evolve. The solution is the short-time Fourier transform (STFT), which computes FFTs on short overlapping windows and produces a spectrogram — frequency versus time — showing spectral evolution moment by moment. Options A and D describe real but secondary constraints, not the fundamental conceptual limitation.
Question 3 True / False
The Fourier series of a periodic musical tone provides a complete, lossless description of the waveform — knowing all the harmonic amplitudes and phases exactly determines the original signal.
TTrue
FFalse
Answer: True
The Fourier representation is mathematically complete and invertible. Given all coefficients (amplitudes and phases for every harmonic), the original periodic signal can be reconstructed exactly via the inverse Fourier series. The spectrum is not an approximation but an equivalent representation — it contains the same information as the time-domain waveform, just organized differently. This is the power of Fourier analysis: it converts losslessly between the time domain and the frequency domain.
Question 4 True / False
Raising the pitch of a note while preserving its timbre is equivalent to multiplying most harmonic amplitudes by a constant factor.
TTrue
FFalse
Answer: False
Changing pitch shifts the fundamental frequency and all harmonics proportionally in frequency — each harmonic is still at an integer multiple of the new (higher) fundamental, so all frequency values shift upward. But the amplitudes are not multiplied; the relative pattern of harmonic amplitudes (the spectral envelope) remains similar, which is what preserves timbre. Scaling amplitudes would change loudness and timbre, not pitch. Pitch corresponds to fundamental frequency; timbre corresponds to the pattern of relative harmonic amplitudes. These are independent dimensions.
Question 5 Short Answer
Why is the short-time Fourier transform (STFT) used for musical analysis rather than a single Fourier transform applied to the entire signal? What problem does it solve, and what trade-off does it introduce?
Think about your answer, then reveal below.
Model answer: A single Fourier transform averages spectral content across the entire signal duration, eliminating all time-dependent information. For music — where notes, harmonics, dynamics, and timbres change constantly — this means you cannot determine when particular frequencies are present or how the spectrum evolves. The STFT solves this by applying Fourier transforms to short overlapping windows, producing a spectrogram: a time-frequency representation tracking spectral change moment by moment. The trade-off is the time-frequency uncertainty principle: shorter windows give better time resolution but worse frequency resolution (and vice versa), because you need sufficient signal duration to resolve closely spaced frequencies.
This time-frequency trade-off mirrors the Heisenberg uncertainty principle in quantum mechanics and is a fundamental constraint in signal processing — not a limitation of any particular algorithm. The STFT represents a practical compromise, and choosing window length is an engineering decision that depends on whether time precision or frequency precision matters more for the analysis at hand.