> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://mediakind.ferndocs.com/mkio/reference/encoding-transform-presets/audio-speech-transcription/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://mediakind.ferndocs.com/_mcp/server.

# Transcription transform

MK.IO offers an AI-powered VOD transcription feature that allows you to generate subtitles from audio tracks. The generated subtitles are delivered in WebVTT format into the output asset and can be subsequently added to the manifest of the asset with a [track insertion transform.](/mkio/reference/encoding-transform-presets/add-a-track-to-a-vod-asset)

On-demand transcription is only available for MP4 assets.

MK.IO provides the built-in **Predefined\_ACSVodTranscription** preset for this workflow. When selecting the preset, specify the original language of the audio track using a BCP-47 locale.

## Source language

MK.IO uses Azure AI Speech for transcription. Azure maintains the current [language and voice support table](https://learn.microsoft.com/en-us/azure/ai-services/speech-service/language-support) for speech-to-text locales. Confirm that the locale required by your workflow is available before creating the transform.