TikTok Sound Scraper - Music Metadata & Videos

What is TikTok Sound Scraper - Music Metadata & Videos?

TikTok Sound Scraper is an Apify Actor for people and systems that work from TikTok sound references and need normalized data in the default dataset. It accepts TikTok music URLs, raw clip IDs, or music search queries, and it returns one record per saved sound or one record per saved video using that sound. The most useful fields include clipId, soundUrl, title, author, album, userCount, audioUrl, coverUrl, shareUrl, plus video fields such as videoId, videoUrl, authorUsername, playCount, likeCount, and hashtags. The outcome is a structured dataset for analytics, creator research, trend monitoring, and AI-agent workflows.

This Actor uses ScrapeCreators first and SociaVault as the configured fallback for sound metadata and videos. It is usable through Apify MCP and through the Apify API.

Best fit and connected workflows

This Actor fits workflows that start with a TikTok sound reference and end with a structured list of the sound itself and the videos that use it.

Use it when you want to:

• turn a TikTok /music/ URL into normalized sound metadata,
• resolve a raw clip ID into a sound record,
• discover sounds through TikTok music search and then collect the related sound and video rows,
• export creator and engagement fields for downstream analysis,
• feed sound metadata into another TikTok workflow.

For discovery-first workflows, pass a sound URL or clip ID from a public TikTok video or trend feed into this Actor.

Practical scenario

Maya is reviewing a campaign brief that includes a TikTok music link. She starts with the sound URL https://www.tiktok.com/music/luther-7439295283975702544 and runs the Actor with outputMode set to sounds_and_videos.

The dataset returns a sound row with title, author, album, userCount, audioUrl, and shareUrl, plus video rows with videoId, videoUrl, authorUsername, playCount, likeCount, and caption. Maya uses the userCount and video engagement fields to decide which creators to review next, then opens the listed videoUrl records for a manual content check.

Input fields

Field	Type	Purpose
soundUrls	array of strings	TikTok sound/music URLs
startUrls	array of request sources	Apify-style URL input for TikTok /music/ URLs
clipIds	array of strings	Raw TikTok sound clip IDs
searchQueries	array of strings	Music search queries that use the fallback data provider search
outputMode	string	Save sound metadata, videos, or both
providerOrder	string	scrapecreators-first (default), sociavault-first, scrapecreators-only, or sociavault-only; also available in the Console form
maxSounds	integer	Maximum unique sounds to process
maxSearchResultsPerQuery	integer	Maximum sounds discovered per search query
maxVideosPerSound	integer	Maximum videos saved for each sound
maxPagesPerSound	integer	Safety cap for provider pagination per sound
region	string	Alpha-2 region used for the fallback data provider search
filterBy	string	The fallback data provider search filter: all, title, or creators
sortType	string	The fallback data provider search sort mode
includeRawData	boolean	Include raw primary sound/video objects on each row

Focused input example

{
  "soundUrls": [
    "https://www.tiktok.com/music/luther-7439295283975702544"
  ],
  "outputMode": "sounds_and_videos",
  "maxSounds": 1,
  "maxVideosPerSound": 25,
  "maxPagesPerSound": 5,
  "includeRawData": false
}

Output fields

The default dataset contains normalized sound rows and video rows. Each row includes a recordType and a provider.

Field	Type	Purpose
recordType	string	sound or video
provider	string	Provider that returned the row
inputType	string or null	url, clipId, or search
inputValue	string or null	Original URL, clip ID, or query
searchQuery	string or null	Query that discovered the sound
clipId	string or null	TikTok sound clip ID
soundUrl	string or null	TikTok music page URL
title	string or null	Sound title
author	string or null	Sound author or artist
album	string or null	Album name when available
durationSeconds	number or null	Sound clip duration
fullSongDurationMs	number or null	Matched full song duration
userCount	integer or null	Number of TikTok videos using the sound
createdAt	string or null	Sound or video creation timestamp
isOriginal	boolean or null	TikTok original flag
isOriginalSound	boolean or null	TikTok original audio flag
isCommerceMusic	boolean or null	Commerce music flag
hasCommerceRight	boolean or null	Provider-reported commerce right
canReuse	boolean or null	Reuse or shooting flag when exposed
language	string or null	Reported language
coverUrl	string or null	Cover art or thumbnail URL
audioUrl	string or null	Provider-returned audio preview URL
shareUrl	string or null	Provider-returned TikTok share URL
matchedSongId	string or null	Matched catalog song ID
matchedSongTitle	string or null	Matched catalog song title
matchedSongAuthor	string or null	Matched catalog song author
themeTags	array or null	Genre or mood tags
artists	array or null	Artist profile objects
soundTitle	string or null	Sound title on video rows
soundAuthor	string or null	Sound author on video rows
videoPosition	integer or null	Position in the returned video list
videoId	string or null	TikTok video ID
videoUrl	string or null	TikTok video URL
caption	string or null	Video caption text
authorId	string or null	Video author ID
authorSecUid	string or null	TikTok secure user ID
authorUsername	string or null	Video author username
authorNickname	string or null	Video author display name
authorVerified	boolean or null	Whether the author is verified
region	string or null	Video region when returned
durationMs	integer or null	Video duration in milliseconds
playCount	integer or null	Video play count
likeCount	integer or null	Video like count
commentCount	integer or null	Video comment count
shareCount	integer or null	Video share count
collectCount	integer or null	Video save count
hashtags	array or null	Hashtags detected in the video
scrapedAt	string	Timestamp when the row was saved

Illustrative output record

{
  "recordType": "sound",
  "inputType": "url",
  "inputValue": "https://www.tiktok.com/music/luther-7439295283975702544",
  "clipId": "7439295283975702544",
  "soundUrl": "https://www.tiktok.com/music/luther-7439295283975702544",
  "title": "luther",
  "author": "Kendrick Lamar & SZA",
  "album": "GNX",
  "durationSeconds": 59,
  "userCount": 482194,
  "coverUrl": "https://p16-sg.tiktokcdn.com/aweme/720x720/example.jpeg",
  "audioUrl": "https://sf16-ies-music-sg.tiktokcdn.com/obj/example",
  "shareUrl": "https://www.tiktok.com/music/luther-7439295283975702544",
  "matchedSongTitle": "luther",
  "matchedSongAuthor": "Kendrick Lamar & SZA",
  "scrapedAt": "2026-06-10T00:00:00.000Z"
}

How it works

The Actor accepts direct music URLs, request-list style start URLs, raw clip IDs, and optional search queries.

Implementation facts from the live contract:

• ScrapeCreators is the default first provider for sound details and videos.
• SociaVault is the fallback when the selected primary route fails, including a video pagination HTTP 404. Each provider starts with its own cursor; videos already saved are skipped across pages and providers.
• The fallback data provider search powers searchQueries.
• outputMode selects whether rows include sounds, videos, or both.
• maxSounds, maxSearchResultsPerQuery, maxVideosPerSound, and maxPagesPerSound bound collection size and provider pagination.
• includeRawData adds raw primary sound/video objects to each saved row for debugging and downstream parsing.
• Output rows are stored in the default dataset, and RUN_SUMMARY is stored in the key-value store.

Pricing

This Actor uses Pay per event with Apify platform usage included. The live Pricing tab shows the current pricing details, so check it before each run.

Event examples from the live contract:

• Actor start: $0.00005 per started GB of run memory, rounded up with a minimum of one startup event.
• Scraped TikTok sound: charged for each saved sound metadata row.
• Scraped TikTok video using sound: charged for each saved video row.

For example, two saved sound rows and ten saved video rows cost $0.021 in result events, plus the memory-scaled startup event. Platform usage is included under the current pricing. The live Pricing tab controls the effective charges.

Use with AI agents (MCP)

This Actor is an Apify Actor usable through Apify MCP. It retrieves TikTok sound metadata and video rows from accepted inputs and returns normalized dataset records plus run summary telemetry.

Exact Actor identity: khadinakbar/tiktok-sound-scraper

Use it when an agent needs a structured TikTok music record, a list of videos using a sound, or a sound discovered from search.

Find the TikTok sound behind this URL, save the sound metadata, and return the first 10 videos using it. Use the default provider order, keep raw provider data off, and summarize the sound title, author, usage count, and top creator handles.

Output interpretation:

• recordType = "sound" identifies the music-level row.
• recordType = "video" identifies a video that uses the sound.
• provider shows whether the primary or fallback data provider returned the row.
• scrapedAt marks when the row was saved.
• OUTPUT and RUN_SUMMARY contain the same outcome, provider usage, row counts, cost estimate, stop reason, and per-sound videoCoverage details. An upstream failure, stalled cursor, or page safety cap produces PARTIAL when useful rows were saved. A result cap means the requested sample was collected. Provider exhaustion describes only the returned feed.

Scope and pagination guidance:

• Direct sound URLs and clip IDs map to specific sounds.
• Search queries use the fallback data provider search, then the discovered sounds are scraped.
• maxPagesPerSound is a shared page-request budget across primary and fallback providers for one sound. A fallback restarts from its first page and may consume part of this budget while passing videos already saved. Cursors are managed internally; each bounded execution starts fresh and manages its own video cursors.
• maxVideosPerSound caps the number of saved video rows per sound.
• maxSounds caps the number of unique sounds across direct inputs and search results.

Apify API example

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({
  token: process.env.APIFY_TOKEN,
});

const run = await client.actor('khadinakbar/tiktok-sound-scraper').call({
  soundUrls: ['https://www.tiktok.com/music/luther-7439295283975702544'],
  outputMode: 'sounds_and_videos',
  maxSounds: 1,
  maxVideosPerSound: 10
});

const datasetItems = await client.dataset(run.defaultDatasetId).listItems();

console.log(datasetItems.items);

Best results and outcome guidance

A focused input usually produces the clearest dataset. Start with one of these patterns:

• one sound URL when you already know the exact TikTok music page,
• one clip ID when you only have the numeric sound identifier,
• one search query when you want discovery first and can work from the returned matches,
• sounds mode when you only need sound-level metadata,
• videos mode when you only need the videos that use a sound.

For search-driven workflows, use region, filterBy, and sortType together so the search matches your research context. Set maxVideosPerSound to a smaller number when you only need sample videos, or leave it higher when you want broader coverage of the returned sound.

Continue the workflow

• Then use TikTok Transcript Scraper to continue Tiktok records selected with TikTok Sound Scraper - Music Metadata & Videos into a focused media workflow.
• Then use TikTok Trending Hashtags Scraper to extend TikTok Sound Scraper - Music Metadata & Videos research with a complementary discovery contract.

Design note

I found that the dataset contract cleanly separates sound rows from video rows through recordType, while still keeping shared identifiers like clipId and soundUrl available for both workflow stages.

FAQ

When should I use sound URLs versus clip IDs?

Use sound URLs when you already have a TikTok music page. Use clip IDs when another workflow, such as a search scraper or trend scraper, has already extracted the numeric sound identifier.

How does this Actor connect to discovery workflows?

A discovery Actor can surface public TikTok records first. Pass a sound URL or clip ID from that result into this Actor to collect the related sound metadata and videos.

What appears in the dataset when I only want sound metadata?

Set outputMode to sounds. The dataset will then focus on sound rows, including fields like title, author, album, userCount, audioUrl, and shareUrl.

What appears in the dataset when I only want videos?

Set outputMode to videos. The dataset will focus on video rows with fields such as videoId, videoUrl, authorUsername, playCount, likeCount, and hashtags.

How do search queries work?

searchQueries uses the fallback data provider's TikTok music search endpoint to discover sounds, and then the Actor scrapes each discovered sound according to the other input settings.

Where can I check pricing for the current run?

Open the live Pricing tab in the Actor UI. The Actor uses Pay per event with Apify platform usage included.

Responsible use

Use this Actor for public web data and for workflows that fit your data governance, privacy, and platform-terms requirements. Review the returned dataset fields before downstream use, especially when joining sound rows and video rows into a single analysis table.

Sound video coverage and fallback

The mobile app's post count and a sound's userCount are separate source observations. Neither promises that this API feed can enumerate every public video. Regional availability, ranking, removals, and the upstream endpoint can affect coverage. Increasing the caps allows more pagination; it does not guarantee the app's total. region affects music search only, not direct sound-video enumeration. A 404 during pagination is a provider failure, not proof of exhaustion, and will be visible in providerErrors and videoCoverage.

For a small test with automatic fallback, use:

{
  "clipIds": ["7510997257874753537"],
  "outputMode": "sounds_and_videos",
  "providerOrder": "scrapecreators-first",
  "maxSounds": 1,
  "maxVideosPerSound": 10,
  "maxPagesPerSound": 5,
  "includeRawData": false
}

Use "providerOrder": "sociavault-only" to compare that provider's returned feed. All four choices work on any Apify plan; each saved result uses the existing event pricing and remains subject to the spending limit set for the execution. Choosing another provider does not guarantee more results or matching app counts.

Start another bounded execution to collect more. If an upstream page fails after useful rows were saved, the run reports partial coverage; inspect OUTPUT, RUN_SUMMARY, the dataset, and final charges before starting another bounded run. If a save could not be confirmed, the summary marks the affected totals as unknown rather than claiming zero.

The provider selector controls sound details and videos. Music search is available through SociaVault: automatic orders or SociaVault-only may use it, while ScrapeCreators-only with searchQueries is rejected before a provider request. For a direct sound, region does not change the sound-video feed.

Raw sound output includes the public sound object, rather than the provider account/billing wrapper. Provider account balances are not copied into the Dataset.

TikTok, ScrapeCreators, and SociaVault are trademarks of their respective owners. This independent Actor is not affiliated with, associated with, or endorsed by those services.