Platform links
Paste a public link. We handle extraction, captions and ASR for each platform’s quirks.
API · MCP · Skill - Works everywhere
Transcript API forAudio, Video & Links
Convert audio and video into accurate text with AI-powered transcription. Support 200+ languages, speaker detection, timestamps, subtitles, and AI insights through a simple API.
🎁Start withfree 1,500 API Quota
99.9%
Uptime
24.2s
Response time
24 / 7
Support
#1
Data security
- 200+ Languages
- Direct URL Input
- Speaker Diarization
- Timestamped JSON Output
- AI Chapters & Translation
One API. Every format.
What Is Transcript API?
Built by Video Transcriber AI, Transcript API turns audio, video, and links from YouTube, TikTok, Instagram and more into structured text through one REST endpoint, with speaker diarization, word-level timestamps, and over 200 languages built in.
File upload
Provide a public video or audio file URL — up to 5 GB per file.
- MP4
- MP3
- MOV
- WebM
- AVI
- WAV
- M4A
- AAC
- OGG
Everything runs through the same endpoint — identical JSON envelope, identical error codes, whatever the source. Swap a YouTube link for an MP3 and your code doesn’t change.
Capabilities
What Makes Transcript API Stand Out?
Transcript API is built by Video Transcriber AI to unify every transcription source into a single, developer-friendly endpoint with capabilities that go far beyond basic speech-to-text.
Multi-Platform Source Coverage
Transcript API accepts over ten input types — YouTube, TikTok, Instagram, Facebook, X, BiliBili, Google Drive, Dropbox, and direct file uploads — all through a single endpoint with one response schema.
200+ Language Support
Transcript API covers 200+ languages with automatic detection — no need to specify the source language. Mixed-language audio is handled gracefully, so global teams run one pipeline instead of region-specific models.
Built-in Speaker Diarization
Speaker diarization is included in every Transcript API response at no extra cost. Each segment is tagged with a speaker label, turning multi-person recordings into readable, attributed dialogue.
Timestamped JSON
Transcript API returns clean, machine-readable JSON — every segment stamped with its start time, duration, and text, ready to drop into your pipeline, database, or an LLM’s context. No parsing required.
Multi-Format Export
Transcript API outputs plain TXT, SRT, VTT, and full JSON with metadata. Specify the format in your request and get the transcript already structured for your target system — no conversion scripts needed.
AI chapters & translation
The Transcript API auto-generates chapter markers for easy navigation, plus AI translation between any language pair — turning a single video into structured, multilingual content.
API. MCP. Agent Skills.
How to Use Transcript API?
Getting started with Transcript API takes under five minutes — pick your path: REST API, MCP server, or agent skill. No SDK installation, no infrastructure setup required.
REST API
For developers & automations
Add transcripts to any application. Works with n8n, Make, Zapier, or your own code — any language, any stack.
- Get a free API key at https://videotranscriber.ai/transcript-api
- Make your first API call. Send a POST request to the transcript endpoint with your media URL or file.
- The JSON response includes the full transcript, speaker-labeled segments, and detected language.
MCP server
For ChatGPT, Claude, Cursor...
Paste a video link and your assistant can hear it. It transcribes the audio, then summarizes the key points, answers your questions, and writes notes — all inside the conversation.
claude mcp add --transport http --scope user videotranscriber \
https://videotranscriber.ai/mcp \
--header "Authorization: Bearer $VT_API_KEY"
claude mcp get videotranscriber- Summary
- Meeting Notes
Agent skills
For AI agents
Install once and any AI agent can hear. Summarize videos, turn recordings into notes and docs, or pull code out of a tutorial — whatever your agent already does, now with audio. First use creates an account and grabs an API key. 1500 free Quotas to start.
npx skills add VideoTranscriberAI/videotranscriber-skills \
--skill transcribe-media --agent claude-code --global- Summarize video
- Answer questions
Pricing
Pay for API Quota. No expiration.
One-time packs, no auto-renew, no expiration.
3,000API Quotas
$12
up to 3,000 minutes of transcription
$0.004/API Quota
12,000API Quotas
Save 10%$45$48
up to 12,000 minutes of transcription
$0.004$0.00375/API Quota
Most Popular
30,000API Quotas
Save 20%$96$120
up to 30,000 minutes of transcription
$0.004$0.0032/API Quota
90,000API Quotas
Save 30%$252$360
up to 90,000 minutes of transcription
$0.004$0.0028/API Quota
270,000API Quotas
Save 40%$648$1,080
up to 270,000 minutes of transcription
$0.004$0.0024/API Quota
600,000API Quotas
Save 50%$1,200$2,400
up to 600,000 minutes of transcription
$0.004$0.002/API Quota
Not sure how to integrate?
Need more? Contact UsWhat Users Say About Transcript API
J.P.
Analytics Lead
We tried six transcription APIs and Transcript API is the only one that handles YouTube, TikTok, and file uploads through one endpoint. Speaker diarization works great on podcast recordings — it correctly separated overlapping dialogue better than three competitors we benchmarked. Word-level timestamps let us build an interactive transcript player our users love. Processing is fast even during peak hours, and the documentation saved us at least a week of integration time.
S.N.
Engineering Manager
Transcript API replaced four different tools we were maintaining for our content platform. One endpoint now covers YouTube, TikTok, Instagram, and file uploads — cutting integration code by about 70%. The 200+ language support let us expand into five new markets without adding infrastructure, and automatic detection means non-technical team members can process videos without picking a language.
R.O.
Indie Developer
As a solo developer building a video-to-blog tool, Transcript API was exactly what I needed. One POST endpoint accepts YouTube URLs, TikTok links, or file uploads — no per-platform branching in my code. Speaker diarization is included by default with no extra parameters, perfect for multi-guest podcasts. Word-level timestamps let me auto-generate chapter markers. Credit-based pricing covered my entire MVP build, and scaling was smooth.
M.H.
Legal Tech Lead
We chose Transcript API after a three-week evaluation against AssemblyAI, Deepgram, and Google STT. Multi-source capability plus speaker diarization quality on deposition recordings was the deciding factor — it handled three to five speakers with overlapping dialogue better than any competitor. Word-level timestamps are accurate enough for compliance citation without manual checks. Multi-format export to SRT and VTT streamlined our caption review workflow.
A.P.
EdTech CTO
Transcript API is the backbone of our educational pipeline — 15,000 video lectures per month from YouTube, BiliBili, and direct uploads in 30-plus countries. Automatic language detection saves four hours of manual tagging daily. Speaker diarization on classroom recordings produces clear transcripts our students search and review. Multi-format TXT/SRT/JSON export from one call feeds search, video player, and analytics simultaneously. Costs dropped while quality improved after switching.
L.K.
Data Engineering Director
Our social media monitoring tool tracks brand mentions across YouTube, TikTok, Instagram, Facebook, and X. Transcript API replaced five fragile per-platform scrapers with one API call per video. The response is identical whether the source is a TikTok duet or a YouTube replay. Word-level timestamps surface exact mention moments in long videos, and speaker diarization distinguishes creator speech from background conversation — critical for sentiment accuracy.
Frequently Asked Questions About Transcript API
Transcript API is a multi-source speech-to-text service that converts audio, video, and links from over ten platforms into structured text through one REST endpoint. It includes speaker diarization, word-level timestamps, 200+ language support, and multi-format export without per-feature charges. Developers send one POST request with a URL or file and get back a structured transcript ready for search, analytics, or display.
Transcript API accepts YouTube, TikTok, Instagram, Facebook, X, BiliBili, Google Drive, Dropbox, and direct file uploads through one endpoint. Every source type uses the same request structure and returns the same response schema, so no per-platform client libraries are needed. For YouTube-only workflows, see YouTube to Transcript for a dedicated tool.
Transcript API achieves competitive word error rates using state-of-the-art AI models optimized for real-world audio — background noise, accents, overlapping speakers, and variable recording quality. Speaker diarization runs within the core model, not as post-processing, improving attribution accuracy on multi-person recordings. Test it on your own content with real-world audio samples to evaluate accuracy in your use case.
Transcript API includes comprehensive documentation covering authentication, endpoint reference, request and response schemas, error codes, rate limits, and retry strategies. Code samples are provided in Python, JavaScript, cURL, Go, and Ruby. A live API playground is available so you can test requests against your own media without writing code. Standard REST conventions mean any HTTP client library works without a dedicated SDK.
Transcript API includes speaker diarization in every response at no extra cost. The API identifies and labels individual speakers in multi-person audio — meetings, interviews, podcasts, and panel discussions. Overlapping speech and rapid speaker changes are handled without specifying the number of speakers in advance. Speaker labels are consistent within a session and can be ignored in your parsing if you only need raw text.
Transcript API supports TXT, SRT, VTT, and JSON with full metadata including word-level timestamps, speaker labels, and detected language. Specify the format via a request parameter — the response arrives already structured for your target system. SRT and VTT exports include millisecond-precision timestamps, and JSON exports include all metadata. One request, one format, no conversion scripts required.
Transcript API uses a credit-based pricing model with no free tier. One credit covers one minute of transcription and includes speaker diarization, word-level timestamps, language detection, and subtitle download by default. Chapter summaries and translation features require additional credits. For direct file transcription without API integration, try Transcribe Video to Text.
Transcript API processes media through encrypted channels and does not store uploaded files beyond the time needed for transcription. All API traffic uses TLS, and authentication uses Bearer tokens that can be rotated or revoked. Data processing regions can be specified for compliance requirements. Customer media and transcripts are never used for model training. Detailed security information is available in the privacy policy.
Ship your first transcript in 60 seconds.
Video Transcriber AI Transcript API gives your application everything it needs to convert speech into searchable, structured text — from YouTube clips to uploaded meeting recordings, all through one endpoint with no per-platform setup. 1500 free API Quotas, no credit card, no sales call.
🎁Start withfree 1,500 API Quota