Video Transcriber Ultimate

Transcribe video and audio into accurate, searchable text with timestamps. Paste a Vimeo or SoundCloud URL, or a direct link to an .mp3 or .mp4 file, and get a complete transcript back in seconds. See Supported Platforms for what is verified today.

No API keys needed - AI transcription is included in the actor pricing.

Why Video Transcriber Ultimate?

• Verified Sources - Vimeo, SoundCloud, and any public direct link to an audio or video file, with hundreds more resolvable through yt-dlp
• No API Keys Required - Groq Whisper transcription is included. No need to manage external AI service accounts
• Accurate Timestamps - Get precise start/end times for every segment, perfect for subtitle generation
• 50+ Languages - Automatic language detection or specify from 50+ supported languages
• Rich Metadata - Returns video title, uploader, view count, upload date, and platform information
• Residential Proxies - Optional built-in proxy support for protected platforms (off by default, enable only if you hit blocking)
• Multiple Output Formats - JSON, plain text, or SRT subtitle format

Quick Start

Via Apify Console

{
    "url": "https://player.vimeo.com/video/108650530",
    "language": "auto",
    "includeTimestamps": true
}

Via Apify API (cURL)

curl -X POST "https://api.apify.com/v2/acts/marielise.dev~video-transcriber-ultimate/runs?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"url": "https://player.vimeo.com/video/108650530", "language": "auto"}'

Via Node.js

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: 'YOUR_TOKEN' });

const run = await client.actor('marielise.dev/video-transcriber-ultimate').call({
    url: 'https://player.vimeo.com/video/108650530',
    language: 'auto',
    includeTimestamps: true,
});

const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items[0].text); // Full transcript
console.log(items[0].segments); // Timestamped segments

Via Python

from apify_client import ApifyClient

client = ApifyClient('YOUR_TOKEN')

run = client.actor('marielise.dev/video-transcriber-ultimate').call(run_input={
    'url': 'https://player.vimeo.com/video/108650530',
    'language': 'auto',
    'includeTimestamps': True,
})

items = client.dataset(run['defaultDatasetId']).list_items().items
print(items[0]['text'])  # Full transcript
print(items[0]['segments'])  # Timestamped segments

How It Works

1. Submit a URL - Paste any video URL from a supported platform
2. Download - The actor downloads the video using yt-dlp with residential proxy support
3. Extract Audio - Audio is extracted and optimized for transcription
4. Transcribe - Groq Whisper AI processes the audio with state-of-the-art accuracy
5. Return Results - You receive the full transcript with timestamps and metadata

Supported Platforms

This actor supports transcription from 1000+ video platforms via yt-dlp. Here are some popular ones:

Platform	Status	Notes
Vimeo	Verified	Use the player.vimeo.com/video/ID URL format. Plain vimeo.com/ID links often fail authorization.
SoundCloud	Verified	Podcasts, interviews and any audio-only recording
Direct file URLs	Verified	A public link ending in .mp3 or .mp4, such as a webinar recording
Archive.org	Direct file only	Use archive.org/download/<item>/<file>.mp3. The archive.org/details/<item> page does not download.
TED Talks	Not working	TED extraction is broken upstream. Every TED URL fails during metadata fetch.
YouTube	Not working	The download step requests player clients that YouTube has since walled behind SABR and PO tokens.
Dailymotion	Not working	Metadata resolves, but no standalone audio format is offered, so the download step fails.
Twitch, Rumble, Odysee, Bitchute, PeerTube, Bilibili and others	Unverified	Listed by yt-dlp, but not confirmed against this actor. Platforms that only serve muxed HLS will fail the audio-only download step.

Plus hundreds more listed under yt-dlp supported sites. This actor downloads audio only, so a platform works here only if it exposes a standalone audio stream. A platform that serves video and audio muxed together will resolve its metadata and then fail the download.

Nothing in this table crashes the run. An unsupported or blocked URL finishes successfully with a labelled status record in the key-value store under OUTPUT, and you are not charged.

Platforms Requiring Authentication

Some platforms require authentication cookies to access content:

Platform	Status	Solution
Private videos	Requires cookies	Export cookies from your browser and provide in cookies field
Age-restricted content	Requires cookies	Export cookies from a logged-in browser session
Geo-restricted content	May require proxy	Try different proxy countries in settings

For authenticated access, export cookies from your browser using a cookie exporter extension and paste them in the cookies input field in Netscape format.

Input Parameters

Parameter	Type	Required	Default	Description
url	string	Yes	-	Video URL from any supported platform
language	string	No	"auto"	ISO language code (en, es, ja, etc.) or "auto" for detection
includeTimestamps	boolean	No	true	Word-level timestamps. Segment start/end times are always returned regardless
outputFormat	string	No	"json"	Output format: json, text, or srt
useResidentialProxy	boolean	No	false	Use residential proxies. Off by default; enable only if a platform blocks you
proxyCountry	string	No	"US"	Proxy location (US, GB, DE, etc.)
cookies	string	No	-	Netscape-format cookies for private videos

Language Codes

The actor supports 50+ languages with automatic detection. Common codes:

Code	Language	Code	Language
auto	Auto-detect	ja	Japanese
en	English	zh	Chinese
es	Spanish	ko	Korean
fr	French	ar	Arabic
de	German	ru	Russian
pt	Portuguese	hi	Hindi
it	Italian	nl	Dutch

Output Format

JSON Output (Default)

{
    "url": "https://player.vimeo.com/video/108650530",
    "title": "Try something new for 30 days",
    "duration": 190,
    "language": "English",
    "text": "Full transcript text appears here...",
    "segments": [
        {
            "text": "First segment of speech.",
            "start": 0,
            "end": 4.5
        },
        {
            "text": "Second segment continues here.",
            "start": 4.5,
            "end": 9.2
        }
    ],
    "metadata": {
        "platform": "vimeo",
        "uploader": "Channel Name",
        "viewCount": 15000,
        "uploadDate": "20240115"
    },
    "processedAt": "2025-01-11T12:00:00.000Z",
    "provider": "groq-whisper"
}

Output Fields

Field	Type	Description
url	string	Original input URL
title	string	Video title from platform
duration	number	Video length in seconds
language	string	Detected or specified language
text	string	Complete transcript as single string
segments	array	Array of timestamped text segments
segments[].text	string	Text content of the segment
segments[].start	number	Start time in seconds
segments[].end	number	End time in seconds
metadata	object	Platform-specific metadata
metadata.platform	string	Source platform name
metadata.uploader	string	Channel or uploader name
metadata.viewCount	number	View count (if available)
metadata.uploadDate	string	Upload date in YYYYMMDD format
processedAt	string	ISO timestamp of processing
provider	string	Transcription engine used

SRT Output

When outputFormat is set to "srt", the actor saves an SRT subtitle file to the key-value store:

1
00:00:00,000 --> 00:00:04,500
First segment of speech.

2
00:00:04,500 --> 00:00:09,200
Second segment continues here.

Pricing

Pay Per Event pricing based on video duration. On top of the duration tier, every run is charged $0.10 for the result item written to the dataset, so the real cost of a run is the tier price plus $0.10.

Video Length	Duration Tier	Result Item	Total per Video
Under 5 minutes	$1.25	$0.10	$1.35
5-15 minutes	$1.90	$0.10	$2.00
15-30 minutes	$3.00	$0.10	$3.10
30-60 minutes	$4.50	$0.10	$4.60
60-120 minutes	$7.50	$0.10	$7.60

The minimum charge for a successful run is $1.35. Includes platform fees and AI transcription. No additional API keys or costs required.

Use Cases

Content Repurposing

Convert tutorials, podcasts, and webinars into blog posts, articles, and documentation. Extract key quotes and insights automatically from video content.

Subtitle Generation

Generate accurate SRT/VTT subtitle files with precise timestamps. Perfect for accessibility compliance and international audiences.

Research and Analysis

Transcribe interviews, lectures, and presentations for qualitative research. Make video archives fully searchable and analyzable.

SEO and Content Marketing

Extract transcripts from videos for content gap analysis. Create text-based content from video assets to improve search visibility.

Podcast Show Notes

Generate complete transcripts for podcast episodes. Create timestamped chapter markers and highlight key moments for show notes.

E-Learning Platforms

Create searchable transcripts for educational videos. Enable students to search within video content and jump to specific topics.

Media Monitoring

Monitor and transcribe video content for brand mentions, competitor analysis, or news tracking across multiple platforms.

Limitations

• Maximum video length: roughly 50 minutes in practice. The transcription backend accepts audio uploads up to 25 MB, which is about 50-52 minutes at the bitrate this actor extracts. Videos longer than that fail at the transcription step. The 60-120 minute pricing tier exists but is not reliably reachable today.
• Default run timeout: 300 seconds (5 minutes). Short videos finish comfortably inside it (a 3 minute talk takes about 30 seconds end to end), but longer videos need the run timeout raised in the run options before you start the run.
• TED, YouTube and Dailymotion: currently not working. See Supported Platforms.
• Audio-only download: a platform that does not serve a standalone audio stream cannot be transcribed here.
• Live streams: Not supported (recorded content only)
• DRM-protected content: Cannot be downloaded or transcribed
• Some platforms may require cookies: Private or age-restricted content needs authentication

Platform Notes

Vimeo

Use the player URL format:

https://player.vimeo.com/video/108650530

The standard https://vimeo.com/ID format frequently fails with an authorization error, so prefer the player URL.

Private Videos

For private or age-restricted videos:

1. Log into the platform in your browser
2. Use a cookie exporter extension to export cookies in Netscape format
3. Paste the cookies in the cookies input field

Geo-Restricted Content

If a video is geo-restricted:

1. Enable residential proxy (useResidentialProxy: true - it is off by default)
2. Try different proxy countries in the proxyCountry setting
3. Choose a country where the video is available

FAQ

Do I need my own Whisper API key?

No. Transcription costs are included in the actor pricing. No external accounts or API keys required.

What languages are supported?

Over 50 languages including English, Spanish, French, German, Japanese, Chinese, Korean, Portuguese, Arabic, Hindi, and more. Use "auto" for automatic language detection, which works well for most videos.

How accurate is the transcription?

Groq Whisper AI provides industry-leading accuracy, typically 95%+ for clear audio. Results may vary based on audio quality, background noise, and accents.

Can I transcribe private videos?

Yes, by providing authentication cookies exported from your browser in Netscape format. Use a browser extension like "Get cookies.txt" to export cookies.

What if a platform is blocked?

Enable residential proxies by setting useResidentialProxy to true (it is off by default) and try different proxy countries. The actor uses yt-dlp which supports 1000+ platforms.

How long does transcription take?

Most videos process in under 2 minutes. Longer videos may take proportionally longer depending on duration.

What output formats are available?

The actor supports three output formats:

• JSON (default): Full structured data with metadata
• Text: Plain text transcript saved to key-value store
• SRT: Subtitle file format saved to key-value store

Can I process multiple videos?

The actor processes one video per run. For batch processing, use the Apify API to start multiple runs in parallel or integrate with a scheduler.

Technical Details

• AI Model: Groq Whisper (state-of-the-art speech recognition)
• Video Extraction: yt-dlp (supports 1000+ platforms)
• Proxy Support: Residential proxies with 20+ country options
• Audio Processing: Automatic extraction and optimization
• Runtime: Node.js 20 with TypeScript

---

Built by Ekipnico | Powered by Groq Whisper AI