Speech noise remover
Reduce background noise with local audio filters.
Speech noise remover
Drop your files here
or choose files from your device
د کارولو طریقه Speech noise remover
The speech noise remover cleans background noise from a voice recording using FFmpeg filters in the browser. It applies an 80 Hz high-pass filter to cut rumble, an 8 kHz low-pass filter to cut hiss, and FFT-based denoising, then exports a mono 16 kHz WAV. You can process from 1 to 120 seconds, starting at the beginning.
- Choose a source file. If the tool accepts multiple files, arrange them in the order you want.
- Set maximum seconds.
- Run speech noise remover, review the output, then use the available copy or download controls.
د دې وسیلې وړتیاوې
Uses local FFmpeg signal processing. Noise reduction can affect voice quality; silence cuts can shorten pauses. Review the exported mono WAV.
- Maximum seconds
- Range: 1 to 120
محدودیتونه او پروسس
Image inputs: up to 32 megapixels, batches up to 10 images. Audio/video uploads: 50 MB; transcription: 120 seconds. Edited videos: 20 seconds, 4 fps and 640 pixels wide. Translation: English/Spanish, 30,000 characters; subtitles: 200 cues. Neural speech: 1,000 English characters. Model downloads and memory requirements vary by device.
د بېلګې سرچینه
A useful tool makes everyday work easier. Review the results before sharing.
عامې پوښتنې
What noise does the speech noise remover reduce?
The high-pass filter removes low rumble below 80 Hz, the low-pass filter removes content above 8 kHz, and an FFT denoiser with a -25 dB noise floor reduces steady background noise such as fan or air-conditioning sound.
Why does my cleaned audio sound different?
Output is converted to mono at 16 kHz and frequencies above 8 kHz are cut, which suits speech but removes some brightness. Noise reduction can also affect voice quality, so compare it with the original.
Can I clean the audio from a video file?
Yes. The video stream is ignored and only the audio is processed. The result is a WAV audio file, not a video, covering up to the Maximum seconds value, which defaults to 30.
Where is my input processed?
Processes source in your browser. Copy and download are explicit actions; source is not saved to account history.
What are the input limits?
Image inputs: up to 32 megapixels, batches up to 10 images. Audio/video uploads: 50 MB; transcription: 120 seconds. Edited videos: 20 seconds, 4 fps and 640 pixels wide. Translation: English/Spanish, 30,000 characters; subtitles: 200 cues. Neural speech: 1,000 English characters. Model downloads and memory requirements vary by device.