Vincent Sullivan
Audio Signal Processing Educator | Founder, OnlinePitchDetector.com
Pitch is the most fundamental measurable property of a musical sound. Every note a singer sings, every string a guitarist plucks, every tone a wind player produces has a precise fundamental frequency — a specific number of vibrations per second that determines where it sits in the musical scale. Measuring that frequency accurately, in real time, in a browser, is the problem OnlinePitchDetector.com was built to solve.
My name is Vincent Sullivan. I have spent years researching audio signal processing, with a particular focus on fundamental frequency detection in browser environments — the algorithms, the accuracy variables, the failure modes, and the practical constraints that determine whether a real-time pitch detector is genuinely useful or merely approximate. Most browser-based pitch tools display a note name and call it done. They do not show the frequency in Hz. They do not show the cents deviation from equal temperament. They do not explain what the YIN algorithm is doing or why it sometimes returns an octave error. OnlinePitchDetector.com does all of these things.
The detection pipeline on this site uses the Web Audio API’s AnalyserNode to capture the microphone signal and applies autocorrelation-based fundamental frequency detection — specifically the YIN algorithm principle — to identify F0 in real time. The detected frequency is then mapped to the nearest musical note using the equal temperament formula, and the cents deviation is calculated precisely:
Cents deviation = 1,200 × log₂(f_detected ÷ f_target)
For a detected frequency of 443 Hz against a target A4 of 440 Hz, this gives approximately +11.8 cents — sharp by nearly an eighth of a tone. That level of precision is what professional intonation work requires, and it is what this tool is built to provide. To understand the full detection pipeline, see the How It Works page.
What I Research and Write About
Fundamental Frequency Detection Algorithms The YIN algorithm (de Cheveigné and Kawahara, 2002) remains the standard reference for accurate monophonic F0 detection. It uses autocorrelation to identify the fundamental period of a signal and applies a difference function that reduces the octave errors that affect naive autocorrelation implementations. I research how YIN performs in the browser Web Audio API environment — specifically its accuracy across the full musical range (80–1,100 Hz), its handling of voiced/unvoiced transitions, and its failure modes at the extremes of range and in the presence of noise.
Equal Temperament, Pitch Standards, and Cents Mathematics Equal temperament divides the octave into 12 logarithmically equal semitones. A4 = 440 Hz is the international standard pitch reference (ISO 16). The cents system — 100 cents per semitone, 1,200 cents per octave — provides a perceptually linear scale for expressing pitch deviation. I research and write about the mathematics of equal temperament, the history of pitch standards, and the practical precision requirements for different musical applications. The FAQ covers the most common questions about pitch detection results and what they mean.
Web Audio API Implementation for Real-Time Pitch Detection The Web Audio API provides the tools needed for real-time pitch detection — AudioContext, AnalyserNode, getUserMedia() — but implementing accurate pitch detection within the browser’s scheduling and processing constraints requires careful attention to FFT window size, analysis rate, smoothing constants, and the handling of low-SNR frames. I research and document these implementation choices and their accuracy implications.
Intonation Science and Musical Pitch Perception Accurate pitch detection is useful only when the user understands what the displayed result means in a musical context. I research and write about intonation standards across different musical instruments, the difference between equal temperament and just intonation, the perceptual thresholds for pitch discrimination (approximately 5–6 cents for trained musicians), and what level of cents deviation is musically significant in different performance contexts.
Why I Built This
When I was looking for a browser-based pitch detector that displayed the fundamental frequency in Hz alongside the musical note, showed the cents deviation with sub-cent precision, and explained the algorithm well enough that a developer or musician could understand why it sometimes gets the octave wrong — I found tools that showed a note name and nothing else.
What I built is a pitch detector designed for musicians who care about intonation, developers who need a documented reference implementation, and educators who want to demonstrate pitch science with a transparent tool.
Accuracy Standards
Every algorithm description, accuracy figure, and frequency reference on OnlinePitchDetector.com is cross-referenced against the YIN algorithm paper (de Cheveigné and Kawahara, 2002), the W3C Web Audio API specification, and ISO 16 (acoustic pitch standard, A4 = 440 Hz). Accuracy claims are documented with specific reference to the conditions — frequency range, SNR, microphone quality, OS audio processing — under which they apply. Limitations are stated explicitly and prominently. For how your microphone audio is handled during detection, see the Data Security page.
Tools on This Site
- Online Pitch Detector — onlinepitchdetector.com
Get in Touch
For content corrections, technical questions, or data requests, visit the Contact page or email directly: contact@onlinepitchdetector.com
Response times: technical and content questions within 48–72 hours. Privacy and data requests within 7 business days. GDPR requests within 30 days. CCPA requests within 45 days.
Vincent Sullivan is the founder and sole author of OnlinePitchDetector.com. For a full account of how content is researched and written, see the Editorial Guidelines. Last updated: June 2026.