Written by Vincent Sullivan | Last updated: June 2026
How to Use This Page
This page covers the most common problems encountered when using the Online Pitch Detector — including failed microphone detection, unstable or jumping readings, octave errors, consistent pitch offsets, and mobile-specific issues. Work through the section that matches your situation. If your problem is not covered here, use the Contact page to submit a technical question.
For background on how the detection pipeline works and what a reliable reading looks like, see the How It Works page.
The Tool Will Not Start
Problem: The browser asks for microphone permission but the tool does not respond after I click Allow.
- Another application is holding the microphone. Close Zoom, Teams, Discord, Google Meet, and any other communication application, reload the page, and try again.
- Browser permission was previously blocked. To reset:
- Chrome: Lock icon in address bar → Site settings → Microphone → Allow
- Firefox: Lock icon → Connection secure → More information → Permissions → Use the Microphone → Allow
- Safari: Safari → Settings → Websites → Microphone → find onlinepitchdetector.com → Allow
- System-level microphone access is disabled.
- Windows: Settings → Privacy and Security → Microphone → ensure your browser is listed and allowed
- macOS: System Settings → Privacy and Security → Microphone → ensure your browser is ticked
Problem: The tool loads but shows no pitch reading when I play or sing.
- Microphone input volume is too low. Check your system microphone input level is set to at least 60–70%. On Windows: right-click the speaker icon → Sounds → Recording → select your microphone → Properties → Levels. On macOS: System Settings → Sound → Input → Input volume.
- The wrong input device is selected. Set your intended microphone as the system default input device before loading the page.
- Your tone is too quiet or too breathy. The YIN algorithm requires a clear, voiced sound above the noise floor. Sustain a tone at a confident, supported volume.
- On iOS Safari: Tap the start button clearly — the AudioContext requires a user gesture to unlock before capture begins — then begin playing or singing immediately.
Unstable or Jumping Readings
Problem: The pitch display jumps rapidly between notes.
Run-to-run variation of 2–5 cents is normal for natural tones with slight vibrato or breath support variation. Larger jumps — 50 cents or more between frames — indicate one of the following:
Background noise. Ambient sound in the 80–1,100 Hz range — traffic, HVAC, music, voices — competes directly with the target tone in the YIN difference function. Test in a quieter room or move closer to the microphone.
Not sustaining a consistent pitch. The tool detects instantaneous F0 at each frame. If your pitch moves — through a slide, a wide vibrato, or natural variation in technique — the display will track those changes in real time. For a stable reading, sustain a single pitch at a consistent volume and with minimal vibrato for at least one to two seconds.
OS noise suppression or AGC. Automatic gain control and noise suppression alter the waveform between frames. On Windows, right-click the speaker icon → Sounds → Recording → select your microphone → Properties → Enhancements → tick “Disable all enhancements.” On macOS, noise suppression is applied per-application.
Playing chords or intervals. The YIN algorithm is monophonic. Playing two or more notes simultaneously produces detection instability. Play single notes only.
Octave Errors
Problem: The displayed note is one octave lower than I expect — for example, showing E3 when I am playing E4.
This is the most common non-trivial error in YIN-based pitch detection. It occurs when the algorithm’s difference function peaks at double the true fundamental period (half the true frequency), producing a result one octave too low.
Most common causes:
- Tones with a very strong second harmonic relative to the fundamental — common in certain flute tones in the upper register, certain bowed string tones, and some singing styles that emphasise upper partials
- Low SNR — when the signal is weak relative to background noise, the difference function becomes less clearly defined and secondary minima can become dominant
- Very high frequencies near the top of the detection range (above approximately 800 Hz)
Fixes:
- Play the tone more loudly to improve SNR
- Move the microphone closer to the instrument or voice
- Try a different vowel or articulation — “aah” or “ohh” on voice, different bowing or breathing technique on instruments
- If the reading is consistently exactly one octave low, double the displayed Hz value to get the true F0
Problem: The displayed note is one octave higher than I expect.
Less common than octave-low errors but possible. This can occur when a strong harmonic at twice the fundamental is louder than the fundamental itself and the difference function peaks at the sub-harmonic period rather than the fundamental.
For bass instruments (below approximately 100 Hz) and very low male voices, consumer microphone roll-off at low frequencies attenuates the fundamental relative to harmonics, increasing the likelihood of octave-high errors. An external microphone with better low-frequency response will typically resolve this.
Consistent Pitch Offset
Problem: All my readings are consistently flat or sharp by approximately the same number of cents across all notes.
A systematic offset — the same flat or sharp bias regardless of which note you play — almost always indicates a reference pitch mismatch:
- Your instrument is tuned to a non-standard reference. A4 = 432 Hz gives a consistent flat offset of approximately −31.7 cents. A4 = 415 Hz (baroque pitch) gives approximately −70 cents. A4 = 443 Hz (some European orchestras) gives approximately +11.8 cents.
- Your digital audio workstation or software is set to a non-440 Hz reference. This applies if you are testing audio from a software instrument or DAW-routed signal.
- A capo, alternate tuning, or transposing instrument. If you are testing a B♭ instrument (clarinet, trumpet) that is not transposed, the sounding pitch is a tone lower than the written pitch, producing a consistent offset.
The tool is calibrated to ISO 16 (A4 = 440 Hz). If your application requires a different reference, note the offset and apply it to interpret the readings.
Microphone-Specific Problems
Problem: Results are noticeably better with an external microphone than with my built-in laptop microphone.
This is expected. Built-in laptop microphones have limited frequency response, higher noise floors, and apply more OS-level processing than external USB or condenser microphones. For pitch detection at the low end of the range (below approximately 200 Hz), built-in microphones often produce less reliable results due to low-frequency roll-off. An external microphone rated to 80 Hz or below with a flat frequency response gives substantially more reliable results across the full detection range.
Problem: Results are different between Chrome and Firefox on the same machine.
Chrome applies automatic echo cancellation and noise suppression to microphone input by default, which can alter the waveform and affect YIN accuracy. Firefox handles audio processing differently and may produce a less modified signal. If Chrome gives consistently less accurate results, try Firefox. On Chrome, navigating to chrome://flags and searching for “echo cancellation” or “noise suppression” to disable hardware processing may improve accuracy.
Mobile-Specific Problems
Problem: The pitch detector is less accurate on my phone than on desktop.
Mobile browsers apply additional audio processing (automatic gain control, noise suppression) that can affect pitch detection. Additionally, built-in smartphone microphones, while adequate for speech, have less flat frequency response than dedicated measurement microphones, particularly at the extremes of range. Sustain tones for longer — three to four seconds — and test in a quiet environment. For critical pitch measurement, desktop use with an external microphone is recommended.
Problem: The tool does not detect pitch on my iPhone.
iOS Safari requires a user tap to unlock the AudioContext before microphone capture begins. Tap the start button clearly, then begin playing or singing immediately after. iOS also applies voice processing to microphone input that can affect the YIN difference function for certain instrument tones. If iPhone results are consistently unreliable, use Chrome on Android or a desktop browser.
Still Having Problems?
If none of the above resolves your issue, the FAQ covers additional questions about result interpretation and detection accuracy. For problems not addressed in either page, contact Vincent Sullivan directly via the Contact page with the following information:
- Your browser name and version
- Your operating system and version
- Your microphone type (built-in, USB, condenser, or other)
- Your instrument or voice type
- A description of the result you are seeing and what you expected
- Whether the problem is consistent or intermittent
Response time for technical issues: 24–48 hours.
Written by Vincent Sullivan, founder of OnlinePitchDetector.com. Last updated: June 2026.