How to Use This Page
This page covers the most common problems users encounter when running the Voice Frequency Test — including failed detection, unstable readings, results that seem too high or too low, and microphone permission 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 tool works and what a normal result looks like for your voice type, 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.
This is almost always caused by one of three things:
- Another application is holding the microphone in exclusive mode. Video calling apps — Zoom, Teams, Google Meet, Discord — frequently hold the audio input device exclusively, preventing the browser from opening a new capture stream. Close all other applications that use the microphone completely, reload the page, and try again.
- The browser has a previously blocked permission for this site. If microphone access was blocked on a previous visit, the browser will not re-prompt automatically. To reset permissions:
- Chrome: Click the lock icon in the address bar → Site settings → Microphone → Allow
- Firefox: Click the lock icon → Connection secure → More information → Permissions → Use the Microphone → Allow
- Safari: Safari → Settings → Websites → Microphone → find voicefrequencytest.com → Allow
- A system-level privacy setting is blocking browser microphone access.
- Windows 11/10: Settings → Privacy and Security → Microphone → ensure “Let apps access your microphone” is On and your browser is listed
- macOS: System Settings → Privacy and Security → Microphone → ensure your browser is ticked
Problem: The tool loads but shows no frequency reading when I speak.
The most likely causes are:
- Input volume is too low. Check that your system input volume is set to at least 50–70%. On Windows, right-click the speaker icon in the taskbar → Sounds → Recording → select your microphone → Properties → Levels. On macOS, go to System Settings → Sound → Input → Input volume.
- The wrong microphone is selected as the default input. If you have multiple audio devices connected, the browser may be capturing from a device other than the one you are speaking into. Check your system’s default input device and ensure it matches the microphone you intend to use.
- Whispering or very quiet phonation. The tool requires a clear, voiced sound above the microphone noise floor. Speak or sing at a normal conversational volume or louder.
- The AudioContext was not unlocked by a user gesture on mobile. On iOS Safari, the AudioContext requires a tap on the page before audio capture begins. Tap the start button and then immediately begin speaking.
Unstable or Jumping Readings
Problem: The frequency display jumps rapidly between values — sometimes by 50 Hz or more between frames.
Run-to-run variation of 5–15 Hz is normal in natural speech, where F0 varies continuously with intonation. Variation larger than 20–30 Hz between frames consistently suggests one of the following:
Background noise. Ambient sound with energy in the 80–500 Hz voice frequency range — traffic, HVAC, music, other voices — directly competes with your voice signal in the autocorrelation analysis. The tool cannot reliably distinguish your voice fundamental from background noise at similar frequencies. Test in a quieter environment or move the microphone closer to your mouth.
Breathy or airy phonation. Voices with high breathiness have a less periodic waveform — the vocal folds are not closing fully on each cycle, which introduces aperiodic noise into the signal. This reduces autocorrelation peak clarity significantly. Try supporting the voice with more breath pressure and phonating with a cleaner, more sustained tone.
Inconsistent vocal production. Varying your pitch, volume, or vowel shape rapidly during the analysis window causes the displayed F0 to track those changes in real time. For a stable reading, sustain a single vowel at a single pitch at a consistent volume for at least one to two seconds.
OS noise suppression processing. Windows and macOS apply noise suppression and automatic gain control (AGC) to microphone input by default on many configurations. These processes can introduce signal artefacts that destabilise F0 detection. To disable temporarily on Windows: right-click the speaker icon → Sounds → Recording → select microphone → Properties → Enhancements → tick “Disable all enhancements.” On macOS, noise suppression is applied per-application and may need to be disabled in the calling or audio application settings.
Result Seems Too Low
Problem: My reading is consistently 30–50 Hz lower than I expect for my voice.
Consumer microphone low-frequency roll-off. Built-in laptop microphones and many budget USB microphones have a frequency response that attenuates signals below approximately 100–150 Hz. For male voices with a fundamental in the 80–120 Hz range, this can cause the tool to struggle to detect the true F0 and instead lock onto a stronger harmonic at a higher frequency — or to show a frequency that reflects the attenuated and distorted fundamental.
Fix: Use a microphone with a flat frequency response down to at least 80 Hz. An external USB microphone or a headset microphone positioned close to the mouth typically gives better low-frequency accuracy than a built-in laptop microphone.
Problem: The tool is showing exactly half my expected frequency.
This is a half-period detection error — the autocorrelation algorithm found a lag corresponding to two vocal fold cycles rather than one, producing a result of F0 ÷ 2. This can happen with very consistent, periodic voices where the autocorrelation function has a clear peak at double the true period. If you are expecting approximately 200 Hz and seeing approximately 100 Hz, this is the likely cause. Try humming rather than singing a vowel — the more complex spectral content of a hummed tone sometimes resolves the half-period ambiguity.
Result Seems Too High
Problem: My reading is consistently 30–80 Hz higher than I expect.
Harmonic detection instead of fundamental. In voices where upper harmonics (2×F0, 3×F0) are stronger in amplitude than the true fundamental — common in head voice, falsetto, and some speaking styles — FFT-based analysis may return a harmonic rather than F0. The autocorrelation pipeline reduces this risk significantly, but it is not eliminated entirely with all voice types and phonation styles.
Fix: Try sustaining a lower, more supported vowel sound rather than a bright or thin phonation. Chest voice phonation typically produces a stronger fundamental relative to upper harmonics, making true F0 detection more reliable.
OS audio processing boosting high frequencies. Some system audio configurations apply high-frequency enhancement or equalisation to the microphone input, which can strengthen upper harmonics relative to the fundamental. Check your audio device’s equaliser or enhancement settings and disable any frequency shaping applied to the input.
Microphone-Specific Problems
Problem: The tool works on my laptop but not with my external USB microphone.
Ensure the external USB microphone is set as the default recording device in your system settings before loading the page. Browsers typically capture from the system default input at page load. If you connected the microphone after loading the page, reload the page with the microphone connected and set as default.
Problem: The tool works in Chrome but not in Safari on macOS.
Safari on macOS applies different audio processing to the microphone stream than Chrome. In some configurations, Safari’s default noise suppression and automatic sample rate conversion can introduce processing delay and spectral distortion that affects F0 detection. Try: Safari → Settings → Websites → Microphone → confirm access is set to Allow for voicefrequencytest.com. If the problem persists, Chrome on macOS is recommended for the most consistent results.
Mobile-Specific Problems
Problem: The tool does not detect frequency on my iPhone.
iOS Safari requires a user gesture — a tap — to unlock the AudioContext before audio capture begins. Tap the start button clearly and begin speaking immediately after. Additionally, iOS applies automatic voice processing to microphone input that can interfere with F0 detection for singing voices and sustained tones. For detailed singing or frequency analysis, desktop use is strongly recommended over iOS.
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 Amelia Foster directly via the Contact page with the following information:
- Your browser name and version
- Your operating system and version
- Your microphone type (built-in laptop, USB, headset, or other)
- A description of the result you are seeing or the problem you are experiencing
- Whether the problem is consistent or intermittent
Response time for technical issues: 24–48 hours.
Written by Amelia Foster, founder of VoiceFrequencyTest.com. Last updated: June 2026.