Emotion Recognition with wav2vec2 base on IEMOCAP

This repository provides all the necessary tools to perform emotion recognition with a fine-tuned wav2vec2 (base) model using SpeechBrain. It is trained on IEMOCAP training data.

For a better experience, we encourage you to learn more about SpeechBrain. The model performance on IEMOCAP test set is:

Release	Accuracy(%)
19-10-21	78.7 (Avg: 75.3)

Pipeline description

This system is composed of an wav2vec2 model. It is a combination of convolutional and residual blocks. The embeddings are extracted using attentive statistical pooling. The system is trained with Additive Margin Softmax Loss. Speaker Verification is performed using cosine distance between speaker embeddings.

The system is trained with recordings sampled at 16kHz (single channel). The code will automatically normalize your audio (i.e., resampling + mono channel selection) when calling classify_file if needed.

Install SpeechBrain

First of all, please install the development version of SpeechBrain with the following command:

pip install git+https://github.com/speechbrain/speechbrain.git@$develop

Please notice that we encourage you to read our tutorials and learn more about SpeechBrain.

Perform Emotion recognition

An external py_module_file=custom.py is used as an external Predictor class into this HF repos. We use foreign_class function from speechbrain.pretrained.interfaces that allow you to load you custom model.

from speechbrain.inference.interfaces import foreign_class
classifier = foreign_class(source="speechbrain/emotion-recognition-wav2vec2-IEMOCAP", pymodule_file="custom_interface.py", classname="CustomEncoderWav2vec2Classifier")
out_prob, score, index, text_lab = classifier.classify_file("speechbrain/emotion-recognition-wav2vec2-IEMOCAP/anger.wav")
print(text_lab)

The prediction tensor will contain a tuple of (embedding, id_class, label_name).

Perform Emotion recognition with OpenVINO backend

Unlocking the power of acceleration with Intel OpenVINO runtime as inference "backend", deploy across a mix of Intel® hardware and environments, on-premises and on-device, in the browser or in the cloud. Passing backend as "openvino" in CustomEncoderWav2vec2Classifier class in custom_interface.py, which supports Intel OpenVINO backend to perform model inference with target device.

Steps to perform model inference with OpenVINO

Step 1: Install speechbrain and OpenVINO

First, ensure you have the necessary dependencies installed. Run the following commands to install the development version of SpeechBrain, OpenVINO, and the required version of the transformers library:

pip install git+https://github.com/speechbrain/speechbrain.git@develop --extra-index-url https://download.pytorch.org/whl/cpu
pip install "openvino>=2024.1.0"
pip install "transformers>=4.30.0"

Step 2: Run inference with OpenVINO backend

To run inference using the OpenVINO backend, you can use a sample application. Below is an example script (app.py) demonstrating how to set up and run the model inference:

config = {hints.performance_mode: hints.PerformanceMode.THROUGHPUT}
ov_opts = {"ov_device": "CPU", "config": config}
torch_device = "cpu"

instance = CustomEncoderWav2vec2Classifier(modules=checkpoint.mods,
                                           hparams=hparams_dict, model=classifier.mods["wav2vec2"].model,
                                           audio_file_path="speechbrain/emotion-recognition-wav2vec2-IEMOCAP/anger.wav",
                                           backend="openvino",
                                           opts=ov_opts,
                                           torch_device=torch_device,
                                           save_ov_model=False)

To execute the application, simply run:

python app.py

For more detailed information on optimizing inference with OpenVINO, refer to the OpenVINO Optimization Guide

Inference on GPU

To perform inference on the GPU, add run_opts={"device":"cuda"} when calling the from_hparams method.

Training

The model was trained with SpeechBrain (aa018540). To train it from scratch follows these steps:

Clone SpeechBrain:

git clone https://github.com/speechbrain/speechbrain/

Install it:

cd speechbrain
pip install -r requirements.txt
pip install -e .

Run Training:

cd  recipes/IEMOCAP/emotion_recognition
python train_with_wav2vec2.py hparams/train_with_wav2vec2.yaml --data_folder=your_data_folder

You can find our training results (models, logs, etc) here.

Limitations

The SpeechBrain team does not provide any warranty on the performance achieved by this model when used on other datasets.

Citing SpeechBrain

Please, cite SpeechBrain if you use it for your research or business.

@misc{speechbrain,
  title={{SpeechBrain}: A General-Purpose Speech Toolkit},
  author={Mirco Ravanelli and Titouan Parcollet and Peter Plantinga and Aku Rouhe and Samuele Cornell and Loren Lugosch and Cem Subakan and Nauman Dawalatabad and Abdelwahab Heba and Jianyuan Zhong and Ju-Chieh Chou and Sung-Lin Yeh and Szu-Wei Fu and Chien-Feng Liao and Elena Rastorgueva and François Grondin and William Aris and Hwidong Na and Yan Gao and Renato De Mori and Yoshua Bengio},
  year={2021},
  eprint={2106.04624},
  archivePrefix={arXiv},
  primaryClass={eess.AS},
  note={arXiv:2106.04624}
}

About SpeechBrain

Website: https://speechbrain.github.io/
Code: https://github.com/speechbrain/speechbrain/
HuggingFace: https://huggingface.co./speechbrain/