The speech enhancement SDK for recorded audio.

Studio sound, dereverb, noise removal, and loudness normalization for audio and video files — through Python & JS SDKs or one REST call. No models to ship, no GPUs to run. EU-hosted.

Get API keyRead the docs →
  • Python · JS · REST
  • audio + video in/out
  • EU-hosted · GDPR

One flag per enhancement

Hear it before you build it.

Every demo below is real Cleanvoice output — the same audio your pipeline gets back. Switch versions mid-playback.

client.process(url, studio_sound=True)

Raw recording — dull and uneven

0:00 / 0:00

Same recording — raw vs. studio-enhanced. Switch mid-playback.

This audio is a part of "Podcast 201 // Your body can heal itself with Osteopath Cole Clayton" by School of Calisthenics licensed under CC BY 3.0, edited from original

Raw take in. Studio sound out.

01

Get your API key

Sign up and copy a key from the Cleanvoice dashboard — free credits included.

02

Install the SDK

pip install cleanvoice-sdk or npm i @cleanvoice/cleanvoice-sdk — or call the REST API directly.

03

Send a file, pick your flags

Point it at an audio or video URL and switch on studio_sound, remove_noise, or normalize.

04

Webhook or poll, then download

Jobs run async — SDKs poll for you, or a webhook fires. The result links to your enhanced file.

docs.cleanvoice.ai — full configuration reference →
import { Cleanvoice } from '@cleanvoice/cleanvoice-sdk';const client = new Cleanvoice({  apiKey: process.env.CLEANVOICE_API_KEY,});// EQ, dereverb, leveling, clarity — one flagconst result = await client.process('https://example.com/raw-take.wav', {  studio_sound: true,});console.log('Enhanced:', result.audio.url);

Why developers trust us

All your data in your control. Each file processed securely.

REGION EU-CENTRAL

EU-Hosted

Servers and GPU located in the EU

ISO 27001GDPRAUDITED · ON-PREMISE ON REQUEST

ISO 27001 & GDPR Compliant

DPA, SLA and on request even on premise deployment

YOUR FILESMODEL TRAININGTRAINED ON OUR OWN STUDIO DATA

No Training on Your Data.

We record our own dataset in professional studios.

Simple, transparent pricing

Pay as you go or subscribe. One rate across every capability. Full pricing →

Pay as You Go

  • 5 hours

    $2.20/h

    $11 total

  • 10 hours

    $2.00/h

    $20 total

  • 30 hours

    $1.50/h

    $45 total

  • 200 hours

    $1.00/h

    $200 total

Credits valid for 2 years.

Subscription

  • 10 hours / mo

    $1.10/h

    $11/mo

  • 30 hours / mo

    $1.00/h

    $30/mo

  • 100 hours / mo

    $0.90/h

    $90/mo

  • 200 hours / mo

    $0.88/h

    $175/mo

Unused hours roll over up to 3× your plan.

200+ hours

Custom

From$0.20/ hour

Tailored feature mix and volume pricing for production workloads.

  • Custom API endpoints
  • Priority support
  • Custom billing
Book a call

Uploads are rounded up to the nearest minute.

Enterprise-grade by default

Contractual uptime, audited security, and a team that knows how procurement works.

FORM CV-SLA-01

Uptime SLA

Contractual availability, backed by a public status page.

SIGNED
FORM CV-SUP-02

Dedicated support

A named human, with priority response times.

HUMAN ✓
FORM CV-PRO-03

Procurement-ready

Security questionnaires, DPA review, and vendor onboarding.

FILED
FORM CV-ONB-04

Technical onboarding

Our engineers on the call, from kickoff to production.

SHIPPED
FORM CV-LIC-05

Flexible terms

Pay-as-you-go, subscriptions, or volume agreements.

FAIR
FORM CV-RET-06

Zero retention, optional

Files auto-delete after 7 days by default — or the moment your job finishes.

PURGED

Free credits for baby startups

Working on audio projects for early-stage startup? With a tight budget?

200 hours

Free API credits

Includes all features to edit, enhance, and transcribe.

Write Us and Apply

Write your startup name, what you’re building, and how you’d use Cleanvoice. We will reply within 2 business days.

Frequently askedQuestions.

Ask Now
Is this a real-time speech enhancement SDK?
Today Cleanvoice is file-based: you send recorded audio or video and get the enhanced file back in minutes. Because there is no on-device latency budget, it can run heavier models and combine enhancement with editing in one pass. A real-time SDK for live audio is coming soon — book a call if that is what you need and we will get you early access.
What does studio sound actually do?
studio_sound: true runs the full enhancement chain in one pass: EQ tuned for speech, dereverb, level balancing across speakers, and clarity restoration. remove_noise and normalize are also available as separate flags when you want just one treatment.
Is there an API for Adobe Podcast Enhance?
Adobe Podcast Enhance has no public API — it is a web app you use by hand. Cleanvoice’s studio sound is the same category of speech enhancement, built to be called from code: SDKs for Python and JavaScript, REST for everything else, webhooks when files finish, and volume pricing for pipelines.
Why not self-host DeepFilterNet or RNNoise?
You can — they are good models, and for pure denoising on your own GPUs they may be enough. What you take on is model ops: GPU provisioning, scaling, format handling, and the fact that denoising is only one step. Cleanvoice runs a chain of specialized models (denoise, dereverb, EQ, leveling, plus editing like breath and filler removal) behind one API call, billed per minute of audio instead of per GPU-hour.
Can I enhance and edit in the same request?
Yes — that is the point of the config. One upload can apply studio_sound and normalize while also cutting breaths, fillers, long silences, stutters, and mouth sounds. One result file comes back, with a webhook when it is done.
How do I try it — is there a free tier?
Sign-up comes with free credits, so you can run real files before paying anything. After that it is pay-as-you-go or a subscription, billed by audio length. Early-stage startups can apply for up to 200 hours of free API credits.