Social Video Transcriber

Transform your social videos into actionable insights! Instantly convert Instagram Reels, TikTok, Youtube Shorts and Facebook Reels into searchable transcripts with optional timestamps, speaker labels, and automatic (or manual) language detection. Process one link or many in a single run. Unlock the full value of your content for analysis, accessibility, and growth.

Note: Only publicly available videos can be transcribed. Private or restricted videos are not supported.

Important: Due to CORS security and anti-scraping/bot measures, success rates are not guaranteed. The platform actively blocks automated access, which may result in failed downloads even for public content.

Why Use This Tool?

• Automate tedious work: No more manual transcription—get results in minutes, not hours.
• Unlock hidden insights: Search, analyze, and repurpose your video content for new opportunities.
• Boost accessibility: Make your videos inclusive for all audiences with accurate transcripts.
• Enhance discoverability: Improve SEO and compliance with rich, timestamped text.
• Stay ahead: Track trends, monitor competitors, and optimize your content strategy with ease.

Features

• Supports Instagram Reels, TikTok, Youtube Shorts, Facebook Reels, and Facebook Watch videos (public videos only)
• Bulk transcription: process multiple video URLs in a single run
• Timestamps: [start - end] markers on each line (on by default; can be turned off)
• Optional speaker diarization: label different speakers ([Speaker 0], [Speaker 1], …) — ideal for interviews and podcasts (off by default)
• Language: auto-detect by default, or pick a specific language (see supported languages below)

Use Cases

• Content Creators: Generate captions, repurpose content, and analyze performance.
• Marketers: Track brand mentions, analyze trends, and optimize strategy.
• Researchers: Study social media trends and extract insights from public discourse.
• Businesses: Create training materials, monitor feedback, and maintain compliance records.

Supported Languages

• Bulgarian bg, Catalan ca, Chinese (Simplified) zh-CN, Chinese (Traditional) zh-TW
• Czech cs, Danish da, Dutch / Flemish nl, English en, Estonian et, Finnish fi
• French fr, German de, German (Switzerland) de-CH, Greek el, Hindi hi, Hungarian hu
• Indonesian id, Italian it, Japanese ja, Korean ko, Latvian lv, Lithuanian lt
• Malay ms, Norwegian no, Polish pl, Portuguese pt, Romanian ro, Russian ru
• Slovak sk, Spanish es, Swedish sv, Thai th, Turkish tr, Ukrainian uk, Vietnamese vi

Input Example

{
  "urls": [
    "https://www.instagram.com/reel/example/",
    "https://www.tiktok.com/@user/video/123"
  ],
  "include_timestamps": true,
  "enable_diarization": false,
  "language": "auto"
}

All fields except urls are optional. include_timestamps defaults to true, enable_diarization to false, and language to "auto" (auto-detect).

Output Example (Success)

{
  "sourceUrl": "https://www.instagram.com/reel/example/",
  "videoId": "unique-video-id",
  "status": "success",
  "durationSec": 45.2,
  "transcript": "[0.00s - 2.50s] Welcome to my channel! [2.50s - 8.10s] Today I'm going to show you how to make the perfect pasta dish.",
  "detected_language": "en",
  "speakers_detected": null,
  "timestamp": "2024-01-01T12:00:00.000Z"
}

With include_timestamps: false, the transcript is plain text without the [start - end] markers. With enable_diarization: true, lines are labeled [Speaker 0] ... and speakers_detected reports the speaker count.

Output Example (Failed)

{
  "sourceUrl": "https://www.instagram.com/reel/private/",
  "videoId": "unique-video-id",
  "status": "failed",
  "error": "Failed to download video (video may be private or restricted)",
  "timestamp": "2024-01-01T12:00:00.000Z"
}

Output Fields

• sourceUrl: Original video URL
• videoId: Unique identifier for the video
• status: "success" or "failed"
• durationSec: Video duration in seconds (only for success)
• transcript: Transcribed text — with [start - end] timestamps and/or [Speaker N] labels when those options are enabled (only for success)
• detected_language: BCP-47 code for the detected language (only for success)
• speakers_detected: Number of distinct speakers (only when diarization is enabled; otherwise null)
• error: Error message (only for failed outputs)
• timestamp: Processing time

How to Run on Apify

1. Open the actor on Apify and click Try actor.
2. Add one or more public video URLs in Video URLs (urls). Optionally toggle timestamps, speaker diarization, or pick a language. Private/restricted videos will fail.
3. Start the run. When it finishes, check the Dataset tab for the transcripts.

Input Schema

• urls (string[]): public video URLs to process (one or many).
• include_timestamps (boolean, default true): add [start - end] markers to each line.
• enable_diarization (boolean, default false): label different speakers.
• language (string, default "auto"): spoken language, or auto-detect.
• start_urls (string): deprecated single-URL field, kept for backward compatibility. Ignored when urls is provided.

Output Location

• Results are saved to the run’s Dataset in JSON items matching the examples below.

Usage Limits

• Free tier: 5 runs total. Upgrade to continue.
• Paid plans: Unlimited runs per Apify limits.

Multiple runs require an Apify paid plan. See Apify Pricing.

---

For more details, contact the maintainer or email us at contact@tictech.id