Clean speech from noisy recordings.
Upload a voice recording or pass a URL. Fans, traffic, hum and room noise are removed and you get an MP3 back. Powered by DeepFilterNet.
Built for speech: podcasts, interviews, voice notes, lectures and calls.
Suppresses steady and changing background noise while keeping the voice.
WAV, MP3, M4A, FLAC, OGG or WEBM, any sample rate, up to 10 minutes.
Clean audio returned as MP3.
Extract the audio track, denoise it, then put it back in your editor.
A noisy speech clip and the denoised result.
Authenticate with Authorization: Bearer YOUR_API_KEY (create a key in the dashboard). Poll GET /v1/audio/denoise/{id} until status is completed, then download output_url (MP3).
Multipart form: send either a file or a URL.
| Field | Type | Required | Description |
|---|---|---|---|
audio | file | one of | Audio file (wav/mp3/m4a/flac/ogg/webm), max 10 minutes. |
audio_url | string | one of | Public URL of the audio instead of an upload. |
Billed in credits. 1 credit = $0.001 on the Starter plan; the cash value of a credit depends on your plan. Failed jobs are refunded.
15 credits per minute of input audio (0.25 credits per second, minimum 1 credit). 1 credit = $0.001 on the Starter plan.
Up to 10 minutes per request.
It is designed for speech. For music, use stem separation to isolate parts instead.
DeepFilterNet.
New accounts get 300 free trial credits (one trial per person), no card required.