How to remove a high-pitched ringing without dulling the music
Tell Vizard Agent to remove the ringing specifically and to leave the music alone. A steady whine occupies a narrow band of frequencies, so it can be measured and removed on its own — where a general noise reduction dulls the cymbals, the consonants and everything else on the way past.
What is the short version?
There is a thin electrical whine under the whole clip. It is one tone rather than general noise, and treating it as general noise is exactly what makes the rest of the audio come back muffled — which is why Vizard Agent measures before it filters.
- Tell Vizard Agent to remove the ringing and nothing else.
- Say explicitly that the music must not change.
- Ask it to measure the frequencies before filtering anything.
What do you need before you start?
The file and a description of the sound. Vizard Agent measures the audio itself, but knowing whether it is a steady whine, an intermittent beep or a hum tells it which kind of tone to look for and how narrow the filter can be.
- The file. With its original audio.
- What the sound is like. Whine, hum, beep, hiss.
- Whether it is constant. Or comes and goes.
- What must not change. The music, the voice, the effects.
- How clean it must be. Gone, or just quieter.
What do you type into Vizard Agent?
Name the sound and name the exclusion. "Remove the background noise" is the request that produces a dull result; "remove the high-pitched ringing, do not change the music" is a different job entirely and Vizard Agent treats it as one.
Prompt
Variants worth knowing:
- "Just the ringing." The scope.
- "Do not change the music." The thing a broad filter would damage.
- "Keep the picture untouched." No re-encode of the video.
What does Vizard Agent actually do?
Here is the Vizard Agent sequence on a clip with a steady tone sitting under the soundtrack. The measuring is the entire technique here and the filtering is a single step at the end of it, which is the opposite of how most noise cleanup is done.
- Checks the ringing and targets it while keeping the music intact.
- Locates the unwanted tone and where in the clip it occurs.
- Measures the audio's frequencies across the file.
- Reviews the source before making any change.
- Looks for a narrow ringing frequency so the rest of the soundtrack can stay untouched.
- Finds a steady high-frequency tone and narrows the filter to avoid dulling the music.
- Confirms the narrow noise bands before removing anything.
- Applies the targeted cleanup, leaving the picture unchanged.
- Checks the picture is preserved and compares the cleaned audio against the original.
- Reviews the finished video for audio quality and sync.
- Separates dialogue and music afterwards, to bring the voice forward without touching the track.
- Sets the music lower under speech with gentle lifts between lines, then listens back.
Step five is the difference between this and a noise gate. A filter narrowed to one band removes a tone you can hear and leaves the frequencies either side of it, which is why the music survives.
Step seven is worth insisting on. Confirming the bands before removal is what stops a filter being centred on a harmonic of the tone rather than the tone itself, which leaves the whine in place and takes a chunk of the music with it.
Steps eleven and twelve are the natural follow-on. Once the whine is gone, the dialogue and music can be balanced properly — and doing that after the cleanup rather than before means the balance is set against what you will actually hear.
What does the result look like?
The same soundtrack with the whine missing: the music as bright as it was, the voices as present, the picture untouched and the file the same length. You can hear what changed by comparing the two, and nothing else moved.
When does this not work well?
Some tones cannot be separated. If the whine drifts in pitch, if it sits exactly where an instrument does, or if it is distortion rather than a tone, a narrow filter either misses it or takes the instrument with it.
- Drifting pitch. A fixed filter loses it.
- A tone inside the music. Removing it removes an instrument.
- Clipping and distortion. Not a tone at all.
- Very loud ringing. It is on top of everything, not behind it.
- Intermittent beeps. Better cut than filtered.
How do you fix a result that came back wrong?
Say what you can still hear, or what now sounds wrong that did not before. Vizard Agent can narrow the filter further, move it to a different frequency, or restrict the treatment to only the sections of the clip where the tone actually appears.
- "I can still hear it." The filter is re-centred on the measured frequency.
- "The music sounds dull." The band is narrowed and the depth reduced.
- "It only rings in the first half." The treatment is limited to that range.
- "The voice sounds thin." The dialogue is separated and lifted rather than filtered.
How does Vizard Agent compare to doing it yourself?
By hand you reach for the noise-reduction slider, because it is one control and it does something. It also treats the whole spectrum, and the price of removing a thin whine that way is a soundtrack that sounds like it is playing through a wall.
| By hand | Vizard Agent | |
|---|---|---|
| The tool | One noise-reduction control | A filter measured to the tone |
| The music | Dulled as a side effect | Explicitly protected |
| Verification | Listening | Frequencies measured and compared |
| Scope | The whole file | Only where the tone occurs |
| The mix | Unchanged, or guessed | Rebalanced after the cleanup |
Common questions
What is that ringing usually? Electrical interference from a light or a phone — Vizard Agent treats it the same either way.
Can Vizard Agent hear it without me pointing at it? Often, but naming it saves a pass and avoids a broad filter.
Will the music be exactly the same? Outside the removed band, yes — that is why Vizard Agent narrows the filter.
Can it fix a hum instead? Yes. A mains hum is the same job at a lower frequency.
Does the picture get re-encoded? No. Vizard Agent keeps the video stream untouched.
What if the tone comes and goes? Vizard Agent can treat only the sections where it appears.
Can it also make the voice clearer? Yes, and it is better done after the cleanup.
Will the file get bigger? No. Only the audio track is rewritten.
Can it remove two tones? Yes. Vizard Agent gives each one its own narrow filter.
What if it is in the original recording? That is the usual case, and it is why this is a cleanup rather than a re-record.
How do I check it? Compare Vizard Agent's before and after on headphones, listening to the music rather than the whine.
Can it be done to an audio file alone? Yes. Vizard Agent does not need the video.
Does this work on a podcast? Yes, and a narrow filter is much safer there than broad denoising.
What if the whine is the loudest thing? Then it is unlikely to be recoverable, and Vizard Agent will say so.