emotion2vec (MLX fp16)

MLX (Apple silicon) conversion from the original release. Runs with funasr-mlx (pip install-free repo; see src/funasr_mlx).

Float16 MLX conversion; numerically equivalent to the upstream release within fp16 rounding.

Layout

Flat: config + safetensors at the repository root.

Notices

  • Upstream weights © their respective authors (Apache-2.0); this repository is a numerical conversion and/or quantization derivative. See NOTICE.
  • Conversion/quantization performed with funasr-mlx @ commit 7bb4dc20a4c6.
Downloads last month
19
Safetensors
Model size
0.2B params
Tensor type
F16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support