import onnxruntime as ort
import numpy as np

session = ort.InferenceSession(
    "distilhubert_ser_int8.onnx", providers=["CPUExecutionProvider"]
)

# audio: raw 16 kHz mono waveform, float32, peak-normalized, shape [1, N]
audio = waveform.astype(np.float32)[None, :]

arousal, valence = session.run(["arousal", "valence"], {"audio": audio})
# arousal, valence: float32 [1], each in [-1, 1]

Training

Trained with CCC loss on CREMA-D categorical emotions mapped to dimensional centroids:

ANG → arousal=0.8, valence=-0.6
DIS → arousal=0.3, valence=-0.7
FEA → arousal=0.7, valence=-0.5
HAP → arousal=0.6, valence=0.7
NEU → arousal=0.0, valence=0.0
SAD → arousal=-0.5, valence=-0.4

30 epochs, batch size 16, lr=1e-4, AdamW.

Provenance and License

Full provenance: provenance.json in this repo.

Apache-2.0, matching the ntu-spml/distilhubert backbone this model fine-tunes. (This card previously said MIT; corrected 2026-07-25 to match the backbone's license and the Omote model registry.)

omote-ai/distilhubert-ser

DistilHuBERT SER (Speech Emotion Recognition)

Model Details

Files

Usage (ONNX Runtime)

Training

Provenance and License

omote-ai/
distilhubert-ser