Classify Speech Emotion (HuggingFace)

How it fits

Typical backends

Caveats

• State lifetimeThe model id and the device are watched at runtime and hot-swapped in place when changed, clearing the GPU cache on swap, so switching checkpoints needs no restart.
• FallbackA device set to a CUDA value silently falls back to CPU when no GPU is present, and any CUDA index resolves to GPU 0 rather than the suffix you name.
• AccuracyThe emotion vocabulary and score semantics are defined entirely by the chosen checkpoint, so the label set changes with the model; a lightweight baseline covers only a few coarse classes.
• I/O contractAn empty AudioFrame returns an empty result without running the model, so downstream consumers must handle the empty case.