Skip to content
ModelListedExternal
Download

Distilhubert Ser

Arousal-valence dimensional emotion recognition from speech, built on ntu-spml/distilhubert.

Sourcehuggingface.co/omote-ai/distilhubert-ser/tree/cdab74ee52bd65ef3a9116a9a6868e13042e0d5e
KindOtherVersionRevision cdab74e

Model source

Kind
Other
File size
48.3–90.4 MB
Version
vRevision cdab74e
Updated
Sep 13, 2026
Downloads
0
Source
Hugging Face

Source description

DistilHuBERT SER (Speech Emotion Recognition)

Arousal-valence dimensional emotion recognition from speech, built on ntu-spml/distilhubert.

Model Details

  • Backbone: DistilHuBERT (23.5M params)
  • Head: Linear(768 → 256) → GELU → Dropout(0.3) → Linear(256 → 2) → Tanh
  • Output: arousal ∈ [-1, 1], valence ∈ [-1, 1]
  • Input: Raw 16kHz waveform, variable length
  • Training data: CREMA-D (7,442 samples, 91 speakers)
  • Loss: Concordance Correlation Coefficient (CCC)
  • Best CCC: arousal=0.783, valence=0.737, avg=0.760

Files

FileSizeDescription
distilhubert_ser.onnx90 MBfp32 ONNX model
distilhubert_ser_int8.onnx48 MBINT8 dynamic quantized (deployment)

Usage (ONNX Runtime)

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.)

Sources

1 source
Hugging FacePrimaryVersion cdab74ee52bd65ef3a9116a9a6868e13042e0d5e
huggingface.co/omote-ai/distilhubert-ser/tree/cdab74ee52bd65ef3a9116a9a6868e13042e0d5e

Verified Sep 13

Model artifacts

2 artifacts

distilhubert_ser_int8.onnx

onnx · 48.3 MB · SHA-256 b3bd62c1d1e7…fa97 · Hugging Face

Download

distilhubert_ser.onnx

onnx · 90.4 MB · SHA-256 974f5274ca6b…b485 · Hugging Face

Download

Source excerpts

2 excerpts
ReadmeSep 13huggingface.co/omote-ai/distilhubert-ser

--- license: apache-2.0 tags: - speech-emotion-recognition - onnx - distilhubert - arousal-valence datasets: - CREMA-D language: - en pipeline_tag: audio-classification library_name: onnxruntime --- # DistilHuBERT SER (Speech Emotion Recognition) Arousal-valence dimensional emotion recognition from speech, built on [ntu-spml/distilhubert](https://huggingface.co/ntu-spml/distilhubert). ## Model Details - **Backbone**: DistilHuBERT (23.5M params) - **Head**: Linear(768 → 256) → GELU → Dropout(0.3) → Linear(256 → 2) → Tanh - **Output**: arousal ∈ [-1, 1], valence ∈ [-1, 1] - **Input**: Raw 16kHz waveform, variable length - **Training data**: CREMA-D (7,442 samples, 91 speakers) - **Loss**: Concordance Correlation Coefficient (CCC) - **Best CCC**: arousal=0.783, valence=0.737, avg=0.760 ## Files | File | Size | Description | |------|------|-------------| | `distilhubert_ser.onnx` | 90 MB | fp32 ONNX model | | `distilhubert_ser_int8.onnx` | 48 MB | INT8 dynamic quantized (deployment) | ## Usage (ONNX Runtime) ```python 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](https://huggingface.co/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.)

NoteSep 13huggingface.co/omote-ai/distilhubert-ser

Source context: 0 downloads · 0 likes · Pipeline audio-classification · Library onnxruntime · Repo omote-ai/distilhubert-ser