Previously, the /stream API returned transcoder information separately in fields such as video, audio, global, etc.
At that time, audio-related metadata (e.g. sample_rate, channels, split_channels, volume) was available via the audio field.
Recently, the API response structure has changed:
- video and audio configurations are now merged into the tracks field
However, in the Go SDK, the tracks model does not include important audio metadata, specifically:
- sample_rate
- channels
- split_channels
- volume
At the same time:
- These fields are no longer exposed via the legacy audio field in the SDK
- But the raw Flussonic API response still returns these fields correctly
As a result, the Go SDK currently loses critical audio configuration information, even though it is available from the API.