Extract Audio from Video: MP3, WAV & FLAC in Bulk

This tool lets you extract audio from video files in bulk: give it direct video URLs and it returns a download link for the audio of each one as MP3, WAV, FLAC or M4A. It runs ffmpeg in the cloud, so there is nothing to install and no API key.

What it does not do. It reads direct video files (mp4, mov, webm, mkv, avi, flv) from a public URL. It does not download from YouTube, Vimeo, TikTok, Instagram or other streaming sites and rejects those links without charge. HLS/DASH playlists (.m3u8, .mpd) are not supported. It extracts the first audio track only: it does not transcribe, denoise, normalise volume, split tracks or edit video. A video without sound gives an error row, free of charge. Use it only on videos you own or have the right to process.

What you get

• One dataset row per video: source URL, a download link for the audio file, measured duration, audio minutes billed, size, sample rate, channels, and a status or a clear error.
• Choice of MP3, WAV (16-bit PCM), FLAC (lossless) or M4A (AAC), with bitrate, sample rate (for example 16 kHz for speech-to-text), mono/stereo and an optional start and end time.
• Failed videos are free, never stop the run, and say exactly why they failed.

How to extract audio from a video URL

1. Paste one or more direct video URLs in videoUrls.
2. Pick outputFormat (default mp3) and, if you like, the bitrate, sample rate and channels.
3. Run it. Open the dataset: every row has a url to the audio file. The SUMMARY record counts the videos, the audio minutes billed and lists every failure.

How to convert video to MP3

Leave outputFormat at mp3 and set audioBitrateKbps: 128 is fine for speech, 192 to 320 for music. Sound that the video stores as AAC is decoded and re-encoded to MP3, which is lossy again, so keep the bitrate high if quality matters.

How to get 16 kHz mono WAV for speech-to-text

Set "outputFormat": "wav", "sampleRateHz": 16000 and "channels": "mono". Most transcription models take exactly this format, and the files are small. Feed the links to your transcriber, for example Transcribe Audio to Text: Files, Podcasts & Video URLs.

How to extract only part of a video as audio

Set startSeconds and endSeconds (0 for the end means "until the end of the video"). Billing counts the minutes of the audio actually delivered.

Input

Field	Default	What it does
videoUrls	sample video	Direct video file links (required).
outputFormat	mp3	mp3, wav, flac or m4a.
audioBitrateKbps	192	MP3 and M4A bitrate (32 to 320). Ignored for WAV and FLAC.
sampleRateHz	0 (original)	Resample, for example 16000 or 44100.
channels	original	original, mono or stereo.
startSeconds / endSeconds	0 / 0	Optional time range; 0 end = to the end of the video.
maxVideosPerRun	100	Hard cap on videos per run (max 2000).
maxAudioMinutesPerRun	600	Hard cap on billable minutes; a video that would exceed it is skipped, free.
maxVideoSizeMb	500	Source size cap (max 2048).
maxDurationMinutes	240	Source duration cap.
videoTimeoutSecs	600	Per download and per extraction.
outputStoreName	empty	Save the audio in a named store that is kept until you delete it.

Input example:

{
  "videoUrls": ["https://interactive-examples.mdn.mozilla.net/media/cc0-videos/flower.mp4"],
  "outputFormat": "mp3",
  "audioBitrateKbps": 192
}

Output

From a real local run of the input above:

{
  "sourceUrl": "https://interactive-examples.mdn.mozilla.net/media/cc0-videos/flower.mp4",
  "videoIndex": 1, "format": "mp3", "status": "ok",
  "url": "https://api.apify.com/v2/key-value-stores/<store>/records/flower-001.mp3",
  "key": "flower-001.mp3", "durationSeconds": 5.055, "minutesBilled": 1, "bytes": 122732,
  "audioCodec": "mp3", "sampleRate": 48000, "channels": 2, "bitrateKbps": 192,
  "sourceDurationSeconds": 5.055, "sourceAudioCodec": "aac", "error": null, "note": null
}

A video that cannot be processed looks like this and is not charged:

{
  "sourceUrl": "https://example.com/nope.mp4", "videoIndex": 2, "format": "mp3", "status": "error",
  "url": null, "minutesBilled": 0, "error": "download failed: HTTP 404"
}

On Apify the url is a signed link that downloads without a token. status is ok, error or skipped (spending limit, maxVideosPerRun or maxAudioMinutesPerRun reached).

Use with the Apify API, MCP, n8n and Make

Python:

from apify_client import ApifyClient

client = ApifyClient("<APIFY_TOKEN>")
run = client.actor("data-done-right/extract-audio-from-video").call(run_input={
    "videoUrls": ["https://example.com/webinar.mp4"],
    "outputFormat": "wav", "sampleRateHz": 16000, "channels": "mono"})
for row in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(row["status"], row["url"] or row["error"])

JavaScript:

import { ApifyClient } from 'apify-client';
const client = new ApifyClient({ token: '<APIFY_TOKEN>' });
const run = await client.actor('data-done-right/extract-audio-from-video').call({
  videoUrls: ['https://example.com/webinar.mp4'], outputFormat: 'mp3' });
const { items } = await client.dataset(run.defaultDatasetId).listItems();

In n8n or Make use the Apify node with the actor name and the same JSON. With Apify's MCP server (mcp.apify.com) an AI assistant can call it: "get the audio of this mp4 as MP3". Only a video URL is needed.

Use cases

• Prepare recordings of webinars, meetings or lectures for speech-to-text (16 kHz mono WAV).
• Turn product demos and screen recordings into MP3 for podcasts or audio feeds.
• Archive the sound of your own video library as lossless FLAC.
• Pull the audio of a segment (for example one interview answer) with a start and end time.
• Batch-convert a list of video links from a spreadsheet or dataset into audio files.

How much does it cost

Pay per event: $0.003 per audio minute (audio-minute, $3 per 1,000 minutes). Every started minute of the delivered audio counts, so a 5-second file is billed as 1 minute and a 61-second file as 2. Failed and skipped videos, rejected links and the SUMMARY record are free. There is no start fee. Plan discounts apply (Silver about 10%, Gold and higher about 20%).

• 1,000 minutes (about 17 hours of audio, for example 200 videos of 5 minutes): $3.
• 10,000 minutes: $30.
• 100,000 minutes: $300.

maxAudioMinutesPerRun and maxVideosPerRun cap a run, and if you set a maximum spend per run the Actor stops cleanly when it is reached and removes the file it could not charge for. Apify platform usage is included in this price.

FAQ

Can it extract audio from YouTube videos?

No. YouTube, Vimeo, TikTok, Instagram and other streaming-site links are rejected with a message and no charge. Use a direct link to a video file that you have the rights to.

What if the video has several audio tracks?

Only the first audio track in the file is extracted. Other tracks (for example a second language) are ignored.

Does WAV or FLAC make the audio better quality?

No. They keep what is in the video without further loss, but if the video holds lossy AAC audio the result is still that AAC sound, only stored in a larger file. WAV is big: about 11.5 MB per minute at 48 kHz stereo. Outputs that would be above 700 MB are refused with advice (use MP3, mono, a lower sample rate or a time range).

Which limits apply?

Source files up to 2 GB (default 500 MB) and 4 hours by default, up to 2000 videos per run. Links to private or internal addresses are blocked. The whole video is downloaded once, so a very large file takes time even if you only need a short range.

What happens to a video with no sound?

The row has status error with "the file has no audio track" and it is not charged.

How long are the audio links kept?

Files live in the run's key-value store, which expires with your plan's retention period. Set outputStoreName to keep them in a named store until you delete it.

Related tools

• Video Frame Extractor: Keyframes, Scenes & Thumbnails for AI pulls still images from the same videos.
• Cut Video: Video Trimmer, Clip Cutter & Video to GIF cuts clips before you extract their audio.
• Transcribe Audio to Text: Files, Podcasts & Video URLs turns the audio into a transcript.