Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.sf-voice.sh/llms.txt

Use this file to discover all available pages before exploring further.

​
What we store

We index your recordings and keep the searchable representation: embeddings, transcripts, and metadata. Source files pass through the processing pipeline and are not retained after indexing completes. What we stores:
  • Embeddings (the indexed representation of your media)
  • Asset metadata (asset_id, asset_class, types, custom metadata fields)
  • Ingestion records and timestamps
What we don’t store:
  • Source audio or video files after processing
  • Call content in raw form
  • Transport-level request data after processing completes

​
Regional processing

Data is processed in the region where your account is provisioned.
Customer regionProcessing location
United StatesUnited States
European Union/United KingdomEurope
EU and UK accounts are processed in European infrastructure. If you have a requirement for a specific region not listed here, contact us.

​
EU and UK data handling

For EU and UK customers:
  • Data is processed in European infrastructure.
  • We act as a data processor. You remain the data controller for any personal data in your recordings.
  • You can delete indexed assets through the API. Deletion removes stored representations — embeddings, transcripts, and metadata.
  • Data processing agreements (DPAs) are available on request.

​
Deleting assets

DELETE /v1/assets/:id removes the asset and its indexed representations. The underlying media is not affected because we do not keep the source file after processing. Using the TypeScript SDK: 
await client.deleteAsset("call_001");

​
Questions

Reach us on Discord or Slack for data handling questions, DPA requests, or security documentation.