# Extract Audio

> Pull the audio track out of a video as a standalone MP3 file, to split it into stems, clean the voice or reuse the soundtrack under another clip. No settings.

Source: https://nodaro.ai/docs/nodes/video/extract-audio

The **Extract Audio** node pulls the audio track out of a video and outputs it as a standalone MP3 file. It is the audio half of splitting a clip, and [Remove Audio](https://nodaro.ai/docs/nodes/video/remove-audio) is the video half. The node has no settings: connect a video and run it.

- Found in: Video › Sound for Video
- Output: audio
- API type: `extract-audio`

## When to use it
- Split the sound of a music video into vocals and instrumental with [Audio Separation](https://nodaro.ai/docs/nodes/audio/audio-separation).
- Keep only the voice of a recorded clip with [Voice Extractor](https://nodaro.ai/docs/nodes/audio/voice-extractor), or cut one section out of it with [Trim Audio](https://nodaro.ai/docs/nodes/audio/trim-audio).
- Reuse the soundtrack of a clip under another video, with [Merge Video & Audio](https://nodaro.ai/docs/nodes/video/merge-video-audio).
- Save a clip's sound as a file of its own.

[Transcribe](https://nodaro.ai/docs/nodes/audio/transcribe), [Dubbing](https://nodaro.ai/docs/nodes/audio/dubbing) and [Voice Changer](https://nodaro.ai/docs/nodes/audio/voice-changer) accept a video and read its sound themselves, so they do not need Extract Audio first.

## Quick start
### Add the node

Press Tab on the canvas and choose **Video › Sound for Video › Extract Audio**.

### Connect the video

Wire the clip into the **Video** input.

### Run it

Click **Run** on the node. The MP3 appears on the node, ready for any audio input.

Workflow: The sound of a music video is split into vocals and instrumental, and the instrumental goes under a new scene.

- Upload Video → Extract Audio (video)
- Extract Audio → Audio Separation (audio)
- Audio Separation → Merge Video & Audio (instrumental)
- Generate Video → Merge Video & Audio (media)

## Example

The soundtrack of the boat clip on the [Generate Video](https://nodaro.ai/docs/nodes/video/generate-video#one-start-frame-three-models) page. VEO 3.1 Fast made the sound together with the picture, and Extract Audio returned it as a file of its own, 6 seconds long.

Audio: [The ambient sound of the VEO 3.1 Fast boat clip, extracted.](https://nodaro.ai/docs-media/examples/extract-audio/boat-soundtrack.mp3)

## Inputs
| Input | Accepts | What it does |
| --- | --- | --- |
| **Video** | Video nodes, such as Upload Video, Generate Video and Video URL | The video whose sound you want. Required. |

The output, **Audio**, is the video's sound as an MP3 file.

## Settings
Extract Audio has no settings.

## Troubleshooting
**The run fails with an error.** The video has no audio track, so there is nothing to extract. Check that the clip actually contains sound.

## Tips
- **Get the silent picture instead.** To keep the video and drop its sound, use [Remove Audio](https://nodaro.ai/docs/nodes/video/remove-audio).
- **Put processed sound back.** After you process the extracted sound, combine it with a video again in [Merge Video & Audio](https://nodaro.ai/docs/nodes/video/merge-video-audio).
- **Choose the format or a section.** The output is always MP3. For another format, or to take only part of the sound, use [Trim Audio](https://nodaro.ai/docs/nodes/audio/trim-audio), which can also extract audio from a video.

## Frequently asked questions

### What format is the extracted audio?

Always MP3. To choose another format, or to take only a section of the sound, use Trim Audio, which can also extract audio from a video.

### What happens if my video has no sound?

The run fails with a clear error, because there is no audio track to extract. Check that the clip actually contains sound.

### How do I get the video without its sound?

Use Remove Audio. It is the video half of splitting a clip, and Extract Audio is the audio half.

### Do I need Extract Audio before Transcribe or Dubbing?

No. Transcribe, Dubbing and Voice Changer accept a video and read its sound themselves. Extract Audio is for nodes that take only audio, such as Audio Separation, Voice Extractor and Trim Audio.