MusicReasonLocal Audio Analysis and Recognition
What are the prominent frequencies right here?
Choose a local audio file and a moment in it. This tool measures the most prominent spectral peaks in that one frame, with frequency and level evidence, as a local measurement, not a pitch or source identification.
Answer
Your audio file is analyzed locally and never uploaded, cached, or included in analytics.
Want to learn more? Why Audio Recognition Is Ambiguous
Understand
A spectral peak is a measured local maximum in this one frame's energy, not a claim about which note is "the" pitch or which instrument produced it. A single musical tone often produces several peaks (its harmonics), and a busy mix produces many unrelated peaks at once; this tool reports the evidence without interpreting it further.
Worked example
A clean 1,000 Hz sine tone produces a single dominant peak within about one FFT bin's width of 1,000 Hz. Digital silence at the chosen frame position produces no peaks at all, rather than a false low-level result.
Method and limitations
The tool takes a 2,048-sample frame at your chosen position, applies a Hann window, and computes a windowed FFT. Local maxima are ranked by level and kept only when separated by at least 20 Hz, up to your peak limit. Frequency resolution is fixed by the FFT size and actual sample rate; a peak's true frequency may sit anywhere within one bin's width of the reported value.
Why "prominent" is not the same as "the pitch"
A single played note on a real instrument typically produces a fundamental plus several harmonic peaks, any of which might register as the loudest depending on the instrument's timbre, so the strongest peak is not automatically the fundamental. This measurement reports every prominent peak with its own evidence precisely so you can reason about which one matters for your purpose, rather than having that judgment made for you.