{"uid":"cap_8vxG-ozBmL5yRHHvCiNt_","slug":"youtube-transcript-context-chunker-melchior-labs-042c3513","name":"YouTube Transcript Context Chunker — Melchior Labs","description":"Pay per request for structured YouTube transcripts, lexical passage search, chunks, and timestamp citations over x402.","url":"https://transcript.melchiorlabs.com/v1/transcripts/context?utm_source=zero.xyz","method":"POST","headers":{},"bodySchema":{"type":"object","properties":{"text":{"type":"string","maxLength":1000000,"description":"Plain text, WebVTT, or SRT transcript"},"format":{"enum":["text","vtt","srt"],"type":"string","default":"text"},"segments":{"type":"array","items":{"type":"object","required":["startSeconds","endSeconds","text"],"properties":{"text":{"type":"string","description":"Transcript text spoken during this segment."},"endSeconds":{"type":"number","description":"Segment end time in seconds from the beginning of the source."},"startSeconds":{"type":"number","description":"Segment start time in seconds from the beginning of the source."}}},"maxItems":20000},"sourceUrl":{"type":"string","pattern":"^https?://"},"includeText":{"type":"boolean","default":true},"includeSegments":{"type":"boolean","default":true},"maxChunkCharacters":{"type":"integer","default":2400,"maximum":8000,"minimum":400}}},"responseSchema":{"type":"json","example":{"usage":{"characters":14,"chunkCount":1,"estimatedTokens":4},"chunks":[{"text":"Hello, agents.","index":0,"citation":{"label":"0:00–0:03"},"endSeconds":3,"startSeconds":0}],"requestId":"89416332-bc4e-43da-b0e2-32af08677d9e","transcript":{"text":"Hello, agents.","segmentCount":1}}},"example":null,"exampleRequest":null,"tags":["x402"],"displayCostAmount":"0.003","displayCostAsset":"USDC","priceDynamic":false,"priceHint":null,"priceStatus":"priced","priceSource":"probe","requiresHandshake":false,"reviewCount":0,"rating":{"score":"0.00","successRate":"0.00","reviews":0,"stars":null,"state":"unrated"},"availabilityStatus":"unknown","priceObserved":{"p10Cents":"0.3000","medianCents":"0.3000","p90Cents":"0.3000","minCents":"0.3000","maxCents":"0.3000","p95Cents":"0.3000","sampleCount":1,"varies":false,"failureChargeRate":1},"sessionDeposit":null,"pricing":{"kind":"static","summary":"$0.003/call","primary":{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.003","per":"call","confidence":"exact"},"accepted":[{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.003","per":"call","confidence":"exact"}]},"paymentMethods":[{"uid":"pm_9A0ga_4qObZSOyg11iMGR","protocol":"x402","methodType":"crypto","chain":"base","mode":"charge","costAmount":"0.003","costPer":"request","priority":0,"asset":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","unit":"request","depositMicros":null,"planRef":null}],"brandName":null,"brandSlug":null,"brandBaseUrl":null,"brandDocsUrl":null,"whatItDoes":"Accepts a raw YouTube transcript (text, VTT, or SRT) and returns structured, timestamped chunks with citation labels suitable for RAG, passage search, or agent context injection.","exampleAgentPrompt":"Take this YouTube transcript in VTT format and break it into chunks of up to 2400 characters each, with timestamp citations like '0:00–0:30' so I can use them as cited sources in my RAG pipeline.","exampleUseCases":[{"title":"RAG pipeline over YouTube lectures","prompt":"I have the SRT captions for a 2-hour lecture on machine learning. Can you split them into 2400-character chunks with timestamp labels so I can index them into my vector database and retrieve relevant passages later?"},{"title":"Cited transcript search for podcast agent","prompt":"Here's the plain-text transcript of a podcast episode. Break it into chunks and include the start and end time citations for each one — I want to be able to answer listener questions and point them to the exact timestamp."},{"title":"Video content summarization with context windows","prompt":"I've got a WebVTT transcript from a YouTube explainer video. Chunk it into segments of about 1500 characters with timestamps, and include the full text and segment list so I can feed each chunk to my summarizer separately."}],"resultDescription":"Returns an array of transcript chunks, each with its text content, zero-based index, start/end time in seconds, and a human-readable citation label (e.g. '0:00–0:03'). Also returns aggregated usage stats including total character count, chunk count, and estimated token count, plus a unique requestId and a top-level transcript summary with total segment count and full text.","failureModes":["Missing or empty 'text' or 'segments' field returns a validation error","Transcript exceeding 1,000,000 characters may be rejected or truncated","Invalid format enum value (not 'text', 'vtt', or 'srt') causes a schema validation error","Malformed VTT or SRT input may result in empty or incorrect segment parsing","maxChunkCharacters outside 400–8000 range triggers a validation error","Network or payment authorization failure (x402) returns a 402 response before processing"],"whenToPreferThis":"Choose this endpoint when you have a raw YouTube transcript (in plain text, VTT, or SRT format) and need it converted into structured, timestamped, citation-ready chunks for downstream RAG, passage retrieval, or LLM context injection. It is especially useful when you need precise start/end second metadata and human-readable citation labels per chunk. Prefer it over generic text splitters when working specifically with video transcripts that require temporal grounding.","instructions":null,"reviewSummary":null,"reviewSummaryHighlights":null,"reviewSummaryConcerns":null,"reviewSummaryGeneratedAt":null,"activationCount":0,"lastUsedAt":null,"lastSuccessfullyRanAt":null,"lastHealthCheckAt":"2026-10-24T06:31:56.830Z","isFirstParty":false,"canonicalSlug":"youtube-transcript-context-chunker-melchior-labs-042c3513"}