Skip to content

Introduction ​

The Esfera developer platform converts spatial audio masters into IAMF files from your own code. You send a master over HTTPS, the same encoder that serves the Esfera app does the work, and you download the result. It is one REST API, with a TypeScript SDK and a command-line tool built on top of it, so you can start from whichever fits the job.

What it does ​

A conversion takes one master file and produces one IAMF file (audio/iamf).

  • Sources. An ADM master (WAV, RF64 or BW64 with ADM metadata) or a bare ambisonic WAV of first, second or third order (4, 9 or 16 channels).
  • Presets. A preset decides what happens to the master's objects and which IAMF profile is written. GET /v1/presets lists what the encoder you are calling accepts, and Presets describes each one.
  • Result. An IAMF file you download for 72 hours after the conversion is ready, with loudness, channel and element facts on the conversion record.

How it fits together ​

  1. Create a conversion. You get an id and a presigned URL for the upload.
  2. Upload the master straight to storage. Nothing large passes through the API.
  3. Analyse and encode. With a preset on create, the encode starts by itself after analysis. Without one, you read the analysis and choose.
  4. Follow it by polling, or let a webhook tell you.
  5. Download the IAMF.

Limits you should know ​

LimitValue
Source duration180 minutes per file
Source size100 GiB per file
Output retention72 hours after ready
Request rate120 requests per minute per key
Concurrent conversionsOne per account unless raised for your account

GET /v1/usage returns the limits that apply to your account, so a script can check them before a long upload.

Where to go next ​

developers.esfera.audio · the Esfera Converter API, SDK and CLI