canonical: https://jentic.com/apis/gotranscript.com/gotranscript # GoTranscript Transcription API The GoTranscript Transcription API submits audio and video files for human-graded transcription, AI transcription, proofreading, translation, captioning, subtitling, and forced alignment. Orders are placed by POSTing job parameters to dedicated endpoints (such as /transcriptions, /ai-transcriptions, and /captions) and results are returned via callback URLs. The API also exposes language enumeration and order lookup endpoints so clients can validate input options and check job state. Authentication uses an API token passed as the api_token query parameter. ## For AI agents Submit audio and video files to GoTranscript for human or AI transcription, captioning, subtitling, translation, and forced alignment, and look up order status by ID. ## Scope Does not handle live streaming transcription, real-time speaker diarization, or audio recording capture - use for batch transcription, captioning, and translation jobs only. ## Capabilities - Submit a file URL for human-graded transcription with speaker labels and timestamps - Order AI transcription for fast turnaround at lower cost than human jobs - Generate captions or subtitles in multiple supported languages from a media file - Request a translation of an existing transcription into a target language - Run forced alignment to match an existing transcript to its source audio - List supported languages and capability codes before placing an order - Look up the state and result location of a previously placed order ## Use cases ### Podcast Transcription Pipeline Each new podcast episode is published to a storage URL and a transcription job is opened automatically. POST /transcriptions submits the file and the result is delivered to a callback URL once the human or AI transcript is ready. Turnaround depends on tier - minutes for AI, hours to days for human-graded - but the API surface is the same. Example prompt: Submit https://example.com/episode-42.mp3 for human transcription with English source and request the result at callback URL https://example.com/callbacks/gotranscript. ### Video Captioning at Scale Marketing teams generate burned-in or sidecar captions for video libraries by POSTing each video to /captions or /subtitles. The API accepts the source language and target output format, and returns the deliverable to the callback URL. Useful for pushing accessibility coverage across hundreds of clips without manual upload. Example prompt: Submit a captioning job for https://example.com/clip.mp4 with English source language and request SRT output to a callback URL. ### Multilingual Transcript Translation Once an English transcript is in hand, /translations creates a target-language version without re-transcribing the audio. This is faster and cheaper than transcribing again from the original file in each language. Common pairing: AI transcription first to get the source text, then human-quality translation for the target language. Example prompt: Translate transcription order 555 from English into Spanish and return the order ID for the translation job. ### AI Agent Transcription Service An AI agent built on Jentic accepts a URL or file and decides - based on the user's latency and quality requirements - whether to call AI or human transcription. The agent searches Jentic for 'transcribe an audio file', loads the appropriate GoTranscript operation schema, and submits the order without exposing the api_token. Example prompt: Use Jentic to search 'transcribe an audio file' and submit https://example.com/call.mp3 for AI transcription with English source language. ## Key endpoints | Method | Path | Description | | --- | --- | --- | | POST | `/transcriptions` | Submit a human-graded transcription order | | POST | `/ai-transcriptions` | Submit an AI transcription order | | POST | `/captions` | Generate captions for a media file | | POST | `/subtitles` | Generate subtitles for a media file | | POST | `/translations` | Translate an existing transcription | | POST | `/forced-alignment` | Run forced alignment between audio and a transcript | | GET | `/orders` | List all orders on the account | | GET | `/orders/{orderId}` | Look up a single order by ID | ## Key resources - **Transcription** — Submit human-graded and AI transcription jobs and retrieve order status - **Translation** — Order translation of existing transcripts into target languages - **Captions** — Generate captions and subtitles in supported formats - **Orders** — List orders on the account and look up a single order by ID - **Languages** — Enumerate supported source and target languages ## Why Jentic - **Setup:** Wiring the GoTranscript Transcription API by hand means handling its api_token query-parameter auth and routing each job to the right endpoint for transcription, captions, subtitles, translation, or forced alignment yourself. Through Jentic you install once, import GoTranscript from the API Directory, store the token once, and your agent calls it. - **Permission scoping:** The target of a GoTranscript job travels in the request body rather than the URL path, so limit the agent to the operations it needs, such as submitting a transcription or reading an order, and leave the rest out of the allowed set. You choose the operations it may call, so the agent cannot reach anything you have not included. - **Credential handling:** Your GoTranscript api_token is stored once, encrypted, by your own Jentic One instance and injected into the request at execution time. It never enters the agent's prompt, logs, or context. - **Discovery method:** Agents search Jentic by intent such as 'transcribe an audio file' or 'caption a video', and Jentic returns the matching GoTranscript operation with its input schema so the agent picks the AI or human route by parameter rather than by hand-choosing a URL. ## Related APIs - **AssemblyAI API** — AssemblyAI is an AI-only transcription and speech understanding API - **Deepgram API** — Deepgram is a low-latency AI speech-to-text API with streaming support - **Google Cloud Translation API** — Use Google Translate alongside GoTranscript when fast machine translation of transcripts is enough ## FAQ ### What authentication does the GoTranscript Transcription API use? The GoTranscript API uses an API key passed as the api_token query parameter on each request. When called through Jentic, the api_token is stored encrypted in Jentic's vault and is appended to outgoing requests at execution time so the agent never embeds it in code. ### Can I order both AI transcription and human transcription with the GoTranscript API? Yes. POST /ai-transcriptions submits AI-only jobs for fast, lower-cost output and POST /transcriptions submits human-graded jobs. POST /transcription-proofreading submits an AI draft for human review when you want both speed and accuracy. ### What are the rate limits for the GoTranscript Transcription API? Specific rate limits are not declared in the OpenAPI spec. Order submission is typically modest in volume per account, so a small number of concurrent submissions is expected to be safe. Honor HTTP 429 responses with exponential backoff. ### How do I submit a transcription job through Jentic? Search Jentic for 'transcribe an audio file', load the POST /transcriptions operation, and execute with the file URL, source language, and callback URL in the request body. With pip install jentic, the call is a three-step search, load, execute flow. ### How do I check the status of a GoTranscript order? GET `/orders/{orderId}` returns the state of a single order along with deliverable metadata once the job completes. GET /orders lists orders on the account so an agent can poll for completed jobs without holding individual order IDs. ### Does the GoTranscript API deliver results synchronously or by callback? Results are delivered asynchronously via the callback URL provided in the submission. The POST endpoint returns an order ID immediately, and the deliverable is sent to the callback once transcription, translation, captioning, or alignment completes. ### Can I limit what my agent is allowed to do with the GoTranscript Transcription API? Yes. Because you run Jentic One yourself, your own rules decide which GoTranscript operations and credentials the agent may use, so you add only the operations it needs to the allowed set. For example, you can let it submit a transcription with POST /transcriptions and read status with GET `/orders/{orderId}` while leaving out POST /ai-transcriptions, POST /captions, or POST /translations. Since the job target travels in the request body rather than the URL, scoping happens at the operation level, and the agent cannot call anything you have not included.