Speech Emotion Classification Onnx

--- license: apache-2.0 library_name: onnxruntime base_model: prithivMLmods/Speech-Emotion-Classification tags: - audio-classification - speech-emotion-recognition - wav2vec2 - onnx language: - en pipeline_tag: audio-classification --- # Speech Emotion Classification — ONNX (INT8) ONNX INT8-quantized version of [prithivMLmods/Speech-Emotion-Classification](https://huggingface.co/prithivMLmods/Speech-Emotion-Classification) for on-device inference in macOS apps via ONNX Runtime C API. ## Model Details - **Architecture:** Wav2Vec2ForSequenceClassification (facebook/wav2vec2-base-960h fine-tuned) - **Format:** ONNX INT8 quantized - **Size:** ~91 MB (INT8), ~361 MB (FP32) - **Input:** Raw audio waveform (16kHz, mono), shape `[1, num_samples]` - **Output:** 8-class emotion logits ### Emotion Labels | ID | Label | Full | |----|-------|------| | 0 | ANG | Anger | | 1 | CAL | Calm | | 2 | DIS | Disgust | | 3 | FEA | Fear | | 4 | HAP | Happy | | 5 | NEU | Neutral | | 6 | SAD | Sad | | 7 | SUR | Surprised | ## Usage On-device real-time speech emotion classification. Inference via ONNX Runtime C API. ```swift // Swift — load and run via OnnxRuntimeWrapper let wrapper = OnnxRuntimeWrapper() try wrapper.load(modelPath: "model_int8.onnx") let logits = try wrapper.run(inputName: "input_values", inputData: audioBuffer, inputShape: [1, Int64(audioBuffer.count)]) let emotionIdx = logits.firstIndex(of: logits.max()!)! ``` ## Files - `model_int8.onnx` — INT8 quantized model (recommended for on-device use) - `model.onnx` — FP32 full precision model - `config.json` — Model configuration with label mappings ## Attribution Original model by [prithivMLmods](https://huggingface.co/prithivMLmods). Converted to ONNX by [onnx-community](https://huggingface.co/onnx-community). INT8 quantization and packaging for macOS by [@smkrv](https://huggingface.co/smkrv).

Source context: 0 downloads · 1 likes · Pipeline audio-classification · Library onnxruntime · Repo smkrv/speech-emotion-classification-onnx