A TypeScript library to retrieve transcripts/subtitles from YouTube videos. Supports auto-generated subtitles, multiple languages, and formatting options.
| Requirement | Value |
|---|---|
| Node | >= 18 |
| Module systems | ESM and CommonJS (dual build) |
| Types | Bundled (.d.ts) |
The package ships both an ESM (import) and a CJS (require) entry point from a single install. TypeScript types are included.
# pnpm
pnpm add youtube-transcript-ts
# npm
npm install youtube-transcript-ts
# yarn
yarn add youtube-transcript-tsESM (import):
import { YouTubeTranscriptApi } from 'youtube-transcript-ts';
const api = new YouTubeTranscriptApi();
// Pass a video ID or a full YouTube URL.
const response = await api.fetchTranscript('dQw4w9WgXcQ');
// or: await api.fetchTranscript('https://www.youtube.com/watch?v=dQw4w9WgXcQ');
// Transcript data
console.log(`Found ${response.transcript.snippets.length} lines`);
response.transcript.snippets.slice(0, 3).forEach(snippet => {
console.log(`[${snippet.start.toFixed(1)}s]: ${snippet.text}`);
});
// Video metadata (always included)
console.log(`Title: ${response.metadata.title}`);
console.log(`Author: ${response.metadata.author}`);CommonJS (require):
const { YouTubeTranscriptApi } = require('youtube-transcript-ts');
const api = new YouTubeTranscriptApi();
const response = await api.fetchTranscript('dQw4w9WgXcQ');// Initialize with a comprehensive configuration
const api = new YouTubeTranscriptApi({
// Cache settings
cache: {
enabled: true,
maxAge: 3600000, // 1 hour in milliseconds
},
// Logging settings
logger: {
enabled: true,
namespace: 'transcript-api',
},
// Invidious settings
invidious: {
enabled: false, // disabled by default
instanceUrls: 'https://yewtu.be',
timeout: 10000, // 10 seconds
},
// Proxy settings
proxy: {
enabled: false, // disabled by default
http: 'http://localhost:8080',
https: 'http://localhost:8080',
},
});Invidious proxy support allows you to fetch transcripts even when YouTube API access is restricted or blocked. This feature is particularly useful for:
- Bypassing IP blocks or regional restrictions
- Accessing transcripts without YouTube tracking
- Improving reliability when YouTube API changes
// Setup with Invidious fallback
const api = new YouTubeTranscriptApi({
invidious: {
enabled: true,
instanceUrls: 'https://yewtu.be', // Use a single instance
},
});
// Multiple fallback instances for improved reliability.
// Failover requires more than one distinct host — the list is tried in order.
const apiWithFallbacks = new YouTubeTranscriptApi({
invidious: {
enabled: true,
instanceUrls: ['https://yewtu.be', 'https://invidious.example.org'],
timeout: 8000, // custom timeout in ms
},
});
// Configure Invidious after initialization
const apiLater = new YouTubeTranscriptApi();
apiLater.setInvidiousOptions({
enabled: true,
instanceUrls: 'https://yewtu.be',
});Note for self-hosting Invidious: If you're running your own Invidious instance for transcript fetching, you must set
use_innertube_for_captions: truein your Invidious configuration file for transcript functionality to work properly.
You can configure the library to use HTTP and HTTPS proxies for all outgoing requests. This is useful in environments where direct connections to YouTube are restricted or when you need to route traffic through specific proxy servers.
// Initialize with proxy configuration
const api = new YouTubeTranscriptApi({
proxy: {
enabled: true,
http: 'http://http-proxy-server.com:8080',
https: 'http://https-proxy-server.com:8443',
},
});
// When both HTTP and HTTPS use the same proxy
const apiWithSameProxy = new YouTubeTranscriptApi({
proxy: {
enabled: true,
http: 'http://proxy-server.com:8080',
https: 'http://proxy-server.com:8080',
},
});
// Configure proxy after initialization
const apiLater = new YouTubeTranscriptApi();
apiLater.setProxyOptions({
enabled: true,
http: 'http://username:password@http-proxy.com:8080',
https: 'http://username:password@https-proxy.com:8443',
});
// Disable proxy
apiLater.setProxyOptions({
enabled: false,
});// Get transcript in German, fallback to English
const response = await api.fetchTranscript('VIDEO_ID', { languages: ['de', 'en'] });
console.log(`Language: ${response.transcript.language}`);fetchTranscript takes an options object: { languages?, preserveFormatting?, formatter? }. Available formats: text, json, srt, webvtt.
const textResponse = await api.fetchTranscript('VIDEO_ID', {
languages: ['en'],
formatter: 'text',
});
console.log(textResponse.formattedText); // Plain text stringThe positional form
fetchTranscript(id, languages, preserveFormatting, formatter)is deprecated (boolean trap) and may be removed in a future major. Prefer the options object.
// Set cookies for age-restricted videos
api.setCookies({
CONSENT: 'YES+cb',
VISITOR_INFO1_LIVE: 'your_visitor_info',
});The API throws specific error types for different failure cases. Import the ones you check against:
import {
YouTubeTranscriptApi,
VideoUnavailable,
NoTranscriptFound,
TranscriptsDisabled,
IpBlocked,
} from 'youtube-transcript-ts';
const api = new YouTubeTranscriptApi();
try {
const transcript = await api.fetchTranscript('VIDEO_ID');
} catch (error) {
if (error instanceof VideoUnavailable) {
console.error('Video is not available');
} else if (error instanceof NoTranscriptFound) {
console.error('No transcript found for the requested languages');
} else if (error instanceof TranscriptsDisabled) {
console.error('Transcripts are disabled for this video');
} else if (error instanceof IpBlocked) {
console.error('Your IP address is blocked by YouTube');
} else {
console.error('An unexpected error occurred:', error);
}
}Contributions are welcome! Please feel free to submit a Pull Request.
MIT License - see LICENSE file for details