{ "openapi": "3.0.1", "info": { "title": "Audio & Video Transcriber - Diarization, Subtitles & RAG", "description": "Batch transcribe audio and video URLs or uploads into accurate, timestamped text. Get speaker labels, subtitles, confidence scores, optional sensitive-data redaction, downloadable files, and citation-ready RAG chunks in one Apify run.", "version": "0.1", "x-build-id": "3jRMDxyXJCIOBYuNX" }, "servers": [ { "url": "https://api.apify.com/v2" } ], "paths": { "/acts/fetchfinch~audio-video-transcriber/run-sync-get-dataset-items": { "post": { "operationId": "run-sync-get-dataset-items-fetchfinch-audio-video-transcriber", "x-openai-isConsequential": false, "summary": "Executes an Actor, waits for its completion, and returns Actor's dataset items in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK" } } } }, "/acts/fetchfinch~audio-video-transcriber/runs": { "post": { "operationId": "runs-sync-fetchfinch-audio-video-transcriber", "x-openai-isConsequential": false, "summary": "Executes an Actor and returns information about the initiated run in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK", "content": { "application/json": { "schema": { "$ref": "#/components/schemas/runsResponseSchema" } } } } } } }, "/acts/fetchfinch~audio-video-transcriber/run-sync": { "post": { "operationId": "run-sync-fetchfinch-audio-video-transcriber", "x-openai-isConsequential": false, "summary": "Executes an Actor, waits for completion, and returns the OUTPUT from Key-value store in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK" } } } } }, "components": { "schemas": { "inputSchema": { "type": "object", "properties": { "mediaUrls": { "title": "Direct audio/video URLs", "type": "array", "description": "Public or signed HTTP(S) URLs that return an audio/video file directly. Supported containers include MP3, M4A, WAV, FLAC, OGG, Opus, WebM, MP4, MOV, and MKV.", "default": [], "items": { "type": "string" } }, "mediaFiles": { "title": "Upload audio/video files", "type": "array", "description": "Files uploaded from your computer. Uploads are processed together with Direct audio/video URLs." }, "mode": { "title": "Transcription mode", "enum": [ "standard", "conversation" ], "type": "string", "description": "Standard is best for one primary speaker. Conversation adds versioned speaker diarization for interviews, meetings, podcasts, and calls.", "default": "standard" }, "language": { "title": "Spoken language", "type": "string", "description": "Use auto to detect the dominant language, multi for code-switching, or a BCP-47/ISO language code such as en, de, es, fr, pt-BR, ja, or zh.", "default": "auto" }, "glossary": { "title": "Names and specialist vocabulary", "type": "array", "description": "Up to 100 names, brands, acronyms, product terms, or short phrases whose recognition should be boosted. Nova-3 keyterm charges may apply.", "default": [], "items": { "type": "string" } }, "multichannel": { "title": "Transcribe audio channels separately", "type": "boolean", "description": "Enable when each participant is recorded on a distinct channel, such as stereo call recordings. Do not enable for ordinary joint-stereo music or video.", "default": false }, "redact": { "title": "Sensitive data redaction", "uniqueItems": true, "type": "array", "description": "Optional Deepgram redaction groups. Redaction is applied by the transcription provider and billed as an add-on only after success.", "items": { "type": "string", "enum": [ "pii", "pci", "phi", "numbers", "aggressive_numbers" ] }, "default": [] }, "privacyMode": { "title": "Provider model-improvement opt-out", "type": "boolean", "description": "Request that Deepgram exclude the media from its Model Improvement Program. Keep enabled for production and sensitive recordings.", "default": true }, "outputFormats": { "title": "Downloadable output formats", "uniqueItems": true, "type": "array", "description": "Files written to the run's Key-Value Store after successful transcription.", "items": { "type": "string", "enum": [ "txt", "md", "json", "srt", "vtt" ] }, "default": [ "txt", "md", "json", "srt", "vtt" ] }, "createRagChunks": { "title": "Create citation-ready RAG chunks", "type": "boolean", "description": "Write deterministic chunks with source URL, timestamp range, speaker IDs, token estimate, and content hash to a separate Dataset.", "default": true }, "chunkSizeTokens": { "title": "RAG chunk target tokens", "minimum": 100, "maximum": 2000, "type": "integer", "description": "Approximate target size. Chunks preserve segment boundaries, so actual sizes vary.", "default": 500 }, "chunkOverlapTokens": { "title": "RAG chunk overlap tokens", "minimum": 0, "maximum": 500, "type": "integer", "description": "Approximate overlap between neighboring chunks.", "default": 50 }, "maxMinutesPerItem": { "title": "Maximum minutes per file", "minimum": 1, "maximum": 600, "type": "integer", "description": "Reject longer files before transcription and before minute-based charges.", "default": 240 }, "maxTotalMinutes": { "title": "Maximum minutes for this batch", "minimum": 1, "maximum": 5000, "type": "integer", "description": "Stop scheduling transcription when the cumulative rounded duration would exceed this limit.", "default": 600 }, "maxFileSizeMb": { "title": "Maximum file size in MB", "minimum": 1, "maximum": 2048, "type": "integer", "description": "Reject oversized downloads before they reach the transcription provider.", "default": 1024 }, "maxConcurrency": { "title": "Maximum concurrent files", "minimum": 1, "maximum": 10, "type": "integer", "description": "Concurrent downloads/transcriptions. Three is a safe default while demand and provider limits are being measured.", "default": 3 }, "dryRun": { "title": "Validate and estimate only", "type": "boolean", "description": "Download and inspect each file, then return duration and estimated billable minutes without sending media to Deepgram or charging transcription events.", "default": false }, "storeRawProviderResponse": { "title": "Store raw Deepgram response", "type": "boolean", "description": "Write the full provider JSON to the Key-Value Store. Disabled by default to minimize retained data.", "default": false } } }, "runsResponseSchema": { "type": "object", "properties": { "data": { "type": "object", "properties": { "id": { "type": "string" }, "actId": { "type": "string" }, "userId": { "type": "string" }, "startedAt": { "type": "string", "format": "date-time", "example": "2025-01-08T00:00:00.000Z" }, "finishedAt": { "type": "string", "format": "date-time", "example": "2025-01-08T00:00:00.000Z" }, "status": { "type": "string", "example": "READY" }, "meta": { "type": "object", "properties": { "origin": { "type": "string", "example": "API" }, "userAgent": { "type": "string" } } }, "stats": { "type": "object", "properties": { "inputBodyLen": { "type": "integer", "example": 2000 }, "rebootCount": { "type": "integer", "example": 0 }, "restartCount": { "type": "integer", "example": 0 }, "resurrectCount": { "type": "integer", "example": 0 }, "computeUnits": { "type": "integer", "example": 0 } } }, "options": { "type": "object", "properties": { "build": { "type": "string", "example": "latest" }, "timeoutSecs": { "type": "integer", "example": 300 }, "memoryMbytes": { "type": "integer", "example": 1024 }, "diskMbytes": { "type": "integer", "example": 2048 } } }, "buildId": { "type": "string" }, "defaultKeyValueStoreId": { "type": "string" }, "defaultDatasetId": { "type": "string" }, "defaultRequestQueueId": { "type": "string" }, "buildNumber": { "type": "string", "example": "1.0.0" }, "containerUrl": { "type": "string" }, "usage": { "type": "object", "properties": { "ACTOR_COMPUTE_UNITS": { "type": "integer", "example": 0 }, "DATASET_READS": { "type": "integer", "example": 0 }, "DATASET_WRITES": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_READS": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_WRITES": { "type": "integer", "example": 1 }, "KEY_VALUE_STORE_LISTS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_READS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_WRITES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_INTERNAL_GBYTES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_EXTERNAL_GBYTES": { "type": "integer", "example": 0 }, "PROXY_RESIDENTIAL_TRANSFER_GBYTES": { "type": "integer", "example": 0 }, "PROXY_SERPS": { "type": "integer", "example": 0 } } }, "usageTotalUsd": { "type": "number", "example": 0.00005 }, "usageUsd": { "type": "object", "properties": { "ACTOR_COMPUTE_UNITS": { "type": "integer", "example": 0 }, "DATASET_READS": { "type": "integer", "example": 0 }, "DATASET_WRITES": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_READS": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_WRITES": { "type": "number", "example": 0.00005 }, "KEY_VALUE_STORE_LISTS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_READS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_WRITES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_INTERNAL_GBYTES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_EXTERNAL_GBYTES": { "type": "integer", "example": 0 }, "PROXY_RESIDENTIAL_TRANSFER_GBYTES": { "type": "integer", "example": 0 }, "PROXY_SERPS": { "type": "integer", "example": 0 } } } } } } } } } }