Skip to guide
← Tool guides

Audio, explained · 14 / 36

Speech Denoise

Adapt locally to noise around speech.

Visual walkthrough1 configuration controlsSynced audio walkthrough
See how it works ↓

Strength changes the suppression floor. It shares the 1024-sample latency boundary of the learned denoiser.

01 / Understand

The idea, before the settings.

Speech Denoise adapts locally to background noise around speech without a Learn step. It is a classical signal-processing tool, not cloud inference.

Conversational micSpeech DenoiseCompressor → Output

Use it in AudioRouter

  1. Place Speech Denoise before stages that amplify noise.
  2. Start at Strength 70% and speak normally, including quiet syllables.
  3. Reduce Strength if breathing, consonants or desired ambience become unnatural.

What success looks like

Background is reduced around conversational speech without a manual Learn step.

If it does not work

For music or delicate material, compare Bypass carefully; a speech-oriented cleaner may suppress sound you want to keep.

Select the node to open Properties. Click its name to rename it. While playing, supported parameter-only edits save after a short pause. For structural edits, use Save when the canvas marks a change. Enabled and Bypass are distinct live controls; see the in-app help for their behavior.

02 / Explore

Make the behavior visible.

Synthetic voice-like phrases and hiss demonstrate stronger versus gentler cleanup. The known noise layer is attenuated for teaching; this is not the production speech detector.

The example repeats until you press Stop, and control changes are heard immediately. The picture shows the tool's shape rather than time; follow the timing in Follow the sound below. This is not a DSP measurement or control for your running AudioRouter session.

03 / Configure

Every control, with a reason.

01

Strength

strengthPercent

Amount of adaptive suppression around speech.

Range / choices0 % → 100 %Default70 %

↓ Lower / alternativeRetains more of the original signal.

↑ Higher / alternativePermits stronger attenuation of estimated noise.

Try thisCompare 70% and 40% on quiet speech.

Keep in mind Music and non-speech material may be altered.

04 / Put it to work

Start with a useful example.

Example 1

Start with a conversational microphone

  1. Place Speech Denoise before a Compressor.
  2. Begin at Strength 70%.
  3. Listen to quiet syllables while adjusting strength down or up.
Example 2

Preserve delicate material

  1. For breathing, music or soft voice that sounds thinned, lower Strength.
  2. Use bypass comparison and choose the lowest useful setting.

These are starting points. Compare at a comfortable monitoring level and adjust for your own source.

05 / Get unstuck

Small details that matter.

  • A speech-focused tool may alter music or other wanted non-speech material.
  • Stronger suppression is not automatically clearer speech.