# ElevenLabs Voice Isolator API: access, pricing and code

> Call the official ElevenLabs Voice Isolator API on Runway Dev to strip background noise from speech, with the model ID, price per second, inputs and code.

Author: Runway · Published: 2026-11-04 · Updated: 2026-11-05 · Source: https://dev.runway.com/learn/elevenlabs-voice-isolator-api

## Summary

Runway Dev gives you the official ElevenLabs Voice Isolator API, listed in the Runway Dev catalog as Voice Isolation.\
\
Voice Isolator takes a speech recording and returns the voice alone, with traffic, wind, crowd chatter and room noise taken out.\
\
You call it with the same API key and credit balance as every other model on Runway Dev, from code, a coding agent or a workflow.\
\
This page covers access, pricing, inputs and code, and how your coding agent can build the integration for you.

A noisy take is the usual reason to call it: an interview recorded on the street, a voice memo with a fan behind it, dialogue under wind. One request with the file is all it takes, with no prompt and no settings.

## What is ElevenLabs Voice Isolator?

ElevenLabs Voice Isolator is ElevenLabs' speech isolation model. It takes a recording of someone speaking and returns the same recording with the background removed: the words, the timing and the voice stay, and the noise around them goes. On Runway Dev it runs as `eleven_voice_isolation`, and every request returns the voice as an MP3 of the same length as the recording you sent.

Voice Isolator is built for speech: interviews, podcasts, voiceovers, calls and dialogue tracks pulled from video. Use it to clean a take before transcription, dubbing or a final mix. In the sample below, the second of street noise before the first word comes back as silence, and the voice keeps its timing, so a cleaned dialogue track lines back up under the picture.

![A waveform of a street interview where traffic and crowd noise fill every pause between the words](https://d3phaj0sisr2ct.cloudfront.net/devportal/learn/elevenlabs-voice-isolator-api-noisy-street-c76f6999.mp4 'The recording sent: a fictional street interview with traffic, a bus, horns, footsteps and wind under the voice. Unmute to listen.')

![A waveform of the same interview after isolation, with silence between the phrases](https://d3phaj0sisr2ct.cloudfront.net/devportal/learn/elevenlabs-voice-isolator-api-isolated-street-0561f5fc.mp4 'What Voice Isolator returned: the same voice and timing, with the street taken out. Unmute to listen.')

```text
Hi, I'm Dana, and this is the Corner Table. Today I'm standing outside the Larkspur Street market, where the bakery stall has a line around the block. Let's find out why everyone is waiting for bread at seven in the morning.
```

The voice above was generated with Eleven v3 and mixed over a street ambience from ElevenLabs Sound Effects, both through the Runway Dev API. Neither the speaker nor the place is real. To record a new voiceover from a script instead of cleaning one, [Eleven v4](https://dev.runway.com/learn/eleven-v4-api) generates the speech.

## What are the ways to use ElevenLabs Voice Isolator on Runway Dev?

ElevenLabs Voice Isolator runs on Runway Dev four ways, all from one account:

| Way in                                                                                                                 | Best for                                                                                                    | What you need                                                   |
| ---------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------- |
| [Playground](https://dev.runway.com/endpoints/voice_isolation?modelId=eleven_voice_isolation)                          | Trying a recording, then copying the request as code                                                        | A Runway Dev account                                            |
| [API](https://docs.dev.runwayml.com/guides/using-the-api/) and [SDKs](https://docs.dev.runwayml.com/api-details/sdks/) | Calling Voice Isolator from your own code                                                                   | An API key, plus the Node or Python SDK if you want one         |
| Coding agent                                                                                                           | Having Claude Code, Codex, Cursor or Grok Bot build the API integration into your product, tool or pipeline | The Runway Dev MCP and skills, and an API key for live calls    |
| [Workflows](https://dev.runway.com/workflows)                                                                                                | Voice isolation as one step with other models, run with one API call                                        | A workflow built in the Runway app and published as an endpoint |

## How do you get API access to ElevenLabs Voice Isolator?

You get API access to ElevenLabs Voice Isolator with a Runway Dev account and an API key:

1. Create an account at https://dev.runway.com.
2. Create an API key in your project's dashboard.
3. Add credits to your project's balance.
4. Store the key in your environment as `RUNWAYML_API_SECRET`.
5. Send a request to `POST /v1/voice_isolation` with `"model": "eleven_voice_isolation"` and the `X-Runway-Version` header. The code below is ready to run.

## What is the ElevenLabs Voice Isolator API pricing?

The ElevenLabs Voice Isolator API on Runway Dev costs 1 credit ($0.01) per 6 seconds of input audio, billed on the length of the recording you send.

| Credits | USD |
| --- | --- |
| 1 credit per 6 seconds of input audio | $0.01 per 6 seconds of input audio |

Prices current as of November 5, 2026. 1 credit = $0.01.

How the billing works:

- **You pay for the input's length.** The output is the same length as the recording, so there is one charge per request.
- **Seconds are counted in blocks and rounded up.** A recording that ends partway into a block is billed for the whole block.
- **A request rejected at validation creates no task.** A recording that is too short, too long or in an unsupported format is refused before anything is charged.

Every rate is also listed in the [docs pricing guide](https://docs.dev.runwayml.com/guides/pricing/) and on [Runway Dev pricing](https://dev.runway.com/pricing).

## What inputs does ElevenLabs Voice Isolator accept?

ElevenLabs Voice Isolator takes one input, `audioUri`, the recording to clean. It has no prompt and no settings: the model works out which sound is the voice.

### `POST /v1/voice_isolation`

| Field | Accepts |
| --- | --- |
| `model` (required) | `eleven_voice_isolation` |
| `audioUri` (required) | One audio clip |

Audio goes in as an HTTPS URL, a `runway://` URI from an [upload](https://docs.dev.runwayml.com/assets/uploads) or a data URI; [inputs](https://docs.dev.runwayml.com/assets/inputs) lists sizes and formats. Each table shows what a field accepts on its own. Rules that combine fields are checked when the task is submitted, and the [API reference](https://docs.dev.runwayml.com/api/) has them in full.

What the recording has to be:

- **An audio file.** MP3, WAV, FLAC, M4A or AAC, served with a matching `Content-Type` (see [audio inputs](https://docs.dev.runwayml.com/assets/inputs#audio)). To clean the dialogue of a video, extract its audio track first, for example with `ffmpeg -i clip.mp4 -vn -c:a aac dialogue.m4a`, then put the cleaned track back under the picture.
- **Between 4.6 seconds and one hour long.** The length is read from the file itself when the request comes in.
- **Up to 32 MB by URL.** A data URI carries up to 5 MB. For a larger file, [upload it](https://docs.dev.runwayml.com/assets/uploads), which takes up to 200 MB, and send the `runway://` URI it returns. An hour of speech fits easily as MP3 or M4A.

## How do you use the ElevenLabs Voice Isolator API?

You use the ElevenLabs Voice Isolator API by sending `eleven_voice_isolation` as the model to `POST /v1/voice_isolation` with the recording in `audioUri`. This example request, in curl, Node and Python, cleans the playground's sample recording:

**curl**

```bash
curl -X POST https://api.dev.runwayml.com/v1/voice_isolation \
  -H "Authorization: Bearer $RUNWAYML_API_SECRET" \
  -H "X-Runway-Version: 2024-11-06" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "eleven_voice_isolation",
    "audioUri": "https://d3phaj0sisr2ct.cloudfront.net/devportal/playground-examples/voice-isolation-input.m4a"
  }'

# The response carries the task id. Poll the task until it finishes:
curl https://api.dev.runwayml.com/v1/tasks/$TASK_ID \
  -H "Authorization: Bearer $RUNWAYML_API_SECRET" \
  -H "X-Runway-Version: 2024-11-06"
```

**Node.js**

```javascript
import RunwayML, { TaskFailedError } from '@runwayml/sdk';

// Reads your API key from RUNWAYML_API_SECRET.
const client = new RunwayML();

try {
  const task = await client.voiceIsolation
    .create({
      model: 'eleven_voice_isolation',
      audioUri: 'https://d3phaj0sisr2ct.cloudfront.net/devportal/playground-examples/voice-isolation-input.m4a',
    })
    .waitForTaskOutput();
  console.log(task.output);
} catch (error) {
  if (error instanceof TaskFailedError) {
    console.error(error.taskDetails);
  } else {
    throw error;
  }
}
```

**Python**

```python
from runwayml import RunwayML, TaskFailedError

# Reads your API key from RUNWAYML_API_SECRET.
client = RunwayML()

try:
    task = client.voice_isolation.create(
        model="eleven_voice_isolation",
        audio_uri="https://d3phaj0sisr2ct.cloudfront.net/devportal/playground-examples/voice-isolation-input.m4a",
    ).wait_for_task_output()
    print(task.output)
except TaskFailedError as error:
    print(error.task_details)
```

Isolation is asynchronous:

- **The create request returns a task ID right away**, with an `estimatedCost` in credits, and the audio is processed in the background.
- **Poll `GET /v1/tasks/{id}` with backoff** until the status is `SUCCEEDED` or `FAILED`. `THROTTLED` means the task is waiting in a queue, not rejected.
- **The output is a list holding one MP3 URL.** Output URLs are temporary, so download the file when the task succeeds. See [outputs](https://docs.dev.runwayml.com/assets/outputs).
- **The SDKs poll for you.** `waitForTaskOutput()` and `wait_for_task_output()` raise `TaskFailedError` with the task's details when it fails. For hour-long recordings, pass a longer timeout or keep polling the task ID.

## Why does an ElevenLabs Voice Isolator API request fail?

An ElevenLabs Voice Isolator request that breaks an input rule is rejected with a validation error before a task is made, and a task that fails later carries a failure code that names the cause. The common ones:

- **A validation error on `audioUri` that names `video/mp4`**: the file is a video. Extract the audio track and send that instead.
- **A recording rejected for its duration**: it runs under 4.6 seconds or over an hour. Pad a short clip with silence or split a long one.
- **An `audioUri` rejected for its size**: the file behind the URL is over 32 MB. Upload it and send the `runway://` URI.
- **`ASSET.INVALID`**: the file couldn't be read as audio, or one of its properties is outside what the model takes. Check it against [audio inputs](https://docs.dev.runwayml.com/assets/inputs#audio) and re-export it.
- **`INTERNAL`**: the isolation didn't complete, for example when the model's provider is unavailable. Wait, then retry.
- **An output link that stopped working**: output URLs expire. Download the audio when the task succeeds.

See [task failures](https://docs.dev.runwayml.com/errors/task-failures) for every failure code.

## What does Runway Dev add when you call ElevenLabs Voice Isolator?

Runway Dev serves ElevenLabs Voice Isolator next to [Runway's own models and other labs' models](https://dev.runway.com/models/catalog), behind one API key and one credit balance: you build with a single key and pay a single bill, instead of an account, a key and an invoice per lab.

A coding agent can do the Voice Isolator integration for you on Runway Dev, end to end. Connect the [Runway Dev MCP](https://docs.dev.runwayml.com/guides/mcp/) and the agent works from your live account: the models your project can call with their current prices and input limits, your credit balance and the record behind any task. It writes the integration against what the API accepts today, and when a task fails it looks up the reason and fixes it. The [Runway Dev skills](https://docs.dev.runwayml.com/guides/plugin/) show the agent how to integrate with the official SDKs and how to build with models, Model Routers, Recipes and Workflows.

With [Workflows](https://dev.runway.com/workflows), voice isolation becomes one step of a larger pipeline: extract the audio from a clip, isolate the voice, dub it into another language and put it back under the picture, with [Runway's own models and other labs' models](https://dev.runway.com/models/catalog) in the same graph, and run the whole pipeline with a single API call. Workflows don't need to be built by a developer. A creative or marketing teammate builds and tests one visually in the [Runway app](https://app.runwayml.com/workflows) ([how Workflows work](https://help.runwayml.com/hc/en-us/articles/45763528999699-Introduction-to-Workflows)), then [publishes it as an endpoint](https://help.runwayml.com/hc/en-us/articles/50682960972947-Publishing-a-Workflow-as-an-Endpoint) that your code calls. The [edit and finish video guide](https://dev.runway.com/learn/edit-and-finish-video-api) runs that chain step by step.

## FAQ

**Do I need an ElevenLabs account to use Voice Isolator on Runway Dev?**

A Runway Dev account and API key are all you need to call ElevenLabs Voice Isolator on Runway Dev. Make the key at dev.runway.com, add credits, then send `eleven_voice_isolation` as the model to `POST /v1/voice_isolation`. The usage is billed to your Runway Dev credit balance, with the rest of your models.

**How much does the ElevenLabs Voice Isolator API cost?**

On Runway Dev, ElevenLabs Voice Isolator costs 1 credit ($0.01) per 6 seconds of input audio. You pay for the length of the recording you send. Prices are current as of November 5, 2026.

**Is the Voice Isolator API on Runway Dev the official ElevenLabs model?**

Runway Dev's Voice Isolator API runs the official Voice Isolator model from ElevenLabs. You call it with your Runway Dev API key and the model ID `eleven_voice_isolation`, like every other model on Runway Dev.

**What is the ElevenLabs Voice Isolator model ID on Runway Dev?**

The model ID is `eleven_voice_isolation`, sent to `POST /v1/voice_isolation`. The Runway Dev catalog lists it as Voice Isolation.

**Can ElevenLabs Voice Isolator clean the audio of a video?**

ElevenLabs Voice Isolator cleans a video's dialogue once you send its audio track. Extract the track with ffmpeg or a Workflow's Extract Audio step, isolate the voice, then put the cleaned track back under the picture. A Runway workflow can run all three steps as one API call.

**How long can a recording be for ElevenLabs Voice Isolator?**

A recording for ElevenLabs Voice Isolator on Runway Dev runs between 4.6 seconds and one hour. A URL can point to a file of up to 32 MB, and larger files go in as an upload.

**Can I use ElevenLabs Voice Isolator without writing code?**

You can use ElevenLabs Voice Isolator on Runway Dev without writing code yourself. The playground sends a recording from a form and gives the request back as code, and a coding agent connected to the Runway Dev MCP writes the integration for you. A creative or marketing teammate can also build a workflow with a Voice Isolation step in the Runway app and publish it as an endpoint your app calls.

**Does the same Runway Dev API key work for ElevenLabs Voice Isolator and other models?**

One Runway Dev API key calls ElevenLabs Voice Isolator and every other model your project can use, Runway's own and other labs', with the same SDKs, task flow and credit balance. Switching models means changing the endpoint and the `model` field.

**Can I use ElevenLabs Voice Isolator outputs commercially?**

The [commercial use guide](https://dev.runway.com/learn/commercial-use) covers what you can do with outputs from Runway Dev and where the terms set the rules. You need the rights to the recording you send in.