Clean speech, reduce noise, and optionally reduce room reverb from a public audio URL.

Voice Cleaner API

Use audio-voice-clean for speech cleanup before publishing, transcription, voiceover preparation, or downstream editing. It reduces background noise at three selectable strengths and can additionally reduce room reverb.

Endpoint

POST /api/v1/audio/generations

Poll completion with:

GET /api/v1/tasks/{id}

Example

curl https://reapi.ai/api/v1/audio/generations \
  -H "Authorization: Bearer $REAPI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "audio-voice-clean",
    "audio_url": "https://cdn.example.com/podcast.wav",
    "noise_cancelling_level": 1,
    "encoder_format": "mp3"
  }'

import requests

resp = requests.post(
    "https://reapi.ai/api/v1/audio/generations",
    headers={
        "Authorization": "Bearer rk_live_xxx",
        "Content-Type": "application/json",
    },
    json={
        "model": "audio-voice-clean",
        "audio_url": "https://cdn.example.com/podcast.wav",
        "noise_cancelling_level": 1,
        "encoder_format": "mp3",
    },
    timeout=30,
)
print(resp.json())

const r = await fetch("https://reapi.ai/api/v1/audio/generations", {
  method: "POST",
  headers: {
    Authorization: "Bearer rk_live_xxx",
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    model: "audio-voice-clean",
    audio_url: "https://cdn.example.com/podcast.wav",
    noise_cancelling_level: 1,
    encoder_format: "mp3",
  }),
});
console.log(await r.json());

package main

import (
    "bytes"
    "encoding/json"
    "fmt"
    "io"
    "net/http"
)

func main() {
    body, _ := json.Marshal(map[string]any{
        "model":                  "audio-voice-clean",
        "audio_url":              "https://cdn.example.com/podcast.wav",
        "noise_cancelling_level": 1,
        "encoder_format":         "mp3",
    })
    req, _ := http.NewRequest("POST",
        "https://reapi.ai/api/v1/audio/generations", bytes.NewReader(body))
    req.Header.Set("Authorization", "Bearer rk_live_xxx")
    req.Header.Set("Content-Type", "application/json")

    resp, _ := http.DefaultClient.Do(req)
    defer resp.Body.Close()
    out, _ := io.ReadAll(resp.Body)
    fmt.Println(string(out))
}

Parameters

Field	Type	Required	Notes
`model`	string	yes	Use `audio-voice-clean`.
`audio_url`	URL	yes	Public HTTP(S) source audio URL. Base64 / `data:` URIs are rejected platform-wide.
`noise_cancelling_level`	integer	no	`0`, `1`, or `2` — off, moderate, strongest. Defaults to `0`.
`splitter`	enum	no	`auto`, `andromeda`, `perseus`, `orion`, `phoenix`, `lyra`, or `lynx`. `auto` is omitted upstream so the default engine is used.
`dereverb_enabled`	boolean	no	Additionally reduce room reverb.
`encoder_format`	enum	no	`mp3`, `wav`, `flac`, `aac`, or `ogg`.

Requests are strictly validated: unknown fields are rejected with 400 rather than ignored.

Output

Completed tasks return output.audio_urls and, when available, labeled output.tracks:

{
  "id": "task_018f5a3a1b6e7d9f8c2b4d6e8f0a2c4e",
  "status": "completed",
  "output": {
    "audio_urls": ["https://cdn.reapi.ai/.../clean.mp3"]
  }
}

Billing

Audio jobs are billed per rounded-up source minute — a 2:30 source bills as 3 minutes. Failed jobs refund automatically. See the current per-minute rate on the Voice Cleaner model page.

Which audio model do I need?

Goal	Model	Docs
Denoise / de-reverb speech	`audio-voice-clean`	this page
Keep the vocals, drop the music	`audio-stem-separator` with `stem: "vocals"`	Vocal Remover
Keep the music, drop the vocals	`audio-music-extractor`	Music Extractor
Split several instruments at once	`audio-multistem`	Multistem Splitter
Convert vocals to another voice	`audio-voice-change`	Voice Changer

Voice Cleaner API

Table of Contents