Transcribe Overview

The Transcription endpoint allows you to convert audio and video files into text and SRT subtitles. It uses advanced AI models like Whisper to provide high-quality transcriptions for your media content.

The Transcription

A Transcription represents the process of converting speech to text. It includes the status of the transcription, the time taken to process the file, and once completed, the full transcript text and a download link for the SRT subtitle file.

AI-Powered Transcriptions

FetchMedia leverages state-of-the-art AI technology to ensure that your transcriptions are both fast and accurate. The endpoint supports a wide range of audio and video formats, making it versatile for various media workflows.

• Full Text Transcript: Once a transcription is successful, the complete transcribed text is available directly in the response payload.
• SRT Subtitles: An SRT file is also generated, which can be used to add subtitles to video players or for further editing.

How it Works

When you submit a file for transcription, FetchMedia processes the audio to extract speech and then passes it through our AI models. The process is fully automated and optimized for high-volume workloads.

Input and Output Files

You can provide the input file via a direct file upload or a file_url.

Once the transcription process is complete:

• The transcript field in the transcription object will contain the full text.
• The srt field will provide a URL to download the SRT subtitle file.

The generated SRT file is also stored as a File in your organization's library.

Webhooks

Transcription is an asynchronous process. We recommend providing a webhook URL in your request. FetchMedia will send a POST request to this URL with the full transcription data as soon as the process is complete or if it encounters an error.