by ntu-spml
Open source · 95k downloads · 38 likes
DistilHuBERT is an automatic speech processing model designed to extract efficient audio representations from unlabeled data. It is a lightweight and optimized version of the HuBERT model, reducing its size by 75% while maintaining comparable performance across ten different tasks. Through a multi-task learning distillation approach, it provides a lighter and faster alternative, ideal for resource-constrained applications or on-device deployments. Although it lacks an integrated tokenizer, it can be fine-tuned for tasks like speech recognition by adding a suitable module. The model stands out for its fast training speed and ability to operate with limited data, paving the way for custom or embedded solutions.
DistilHuBERT by NTU Speech Processing & Machine Learning Lab
The base model pretrained on 16kHz sampled speech audio. When using the model make sure that your speech input is also sampled at 16Khz.
Note: This model does not have a tokenizer as it was pretrained on audio alone. In order to use this model speech recognition, a tokenizer should be created and the model should be fine-tuned on labeled text data. Check out this blog for more in-detail explanation of how to fine-tune the model.
Paper: DistilHuBERT: Speech Representation Learning by Layer-wise Distillation of Hidden-unit BERT
Authors: Heng-Jui Chang, Shu-wen Yang, Hung-yi Lee
Abstract Self-supervised speech representation learning methods like wav2vec 2.0 and Hidden-unit BERT (HuBERT) leverage unlabeled speech data for pre-training and offer good representations for numerous speech processing tasks. Despite the success of these methods, they require large memory and high pre-training costs, making them inaccessible for researchers in academia and small companies. Therefore, this paper introduces DistilHuBERT, a novel multi-task learning framework to distill hidden representations from a HuBERT model directly. This method reduces HuBERT's size by 75% and 73% faster while retaining most performance in ten different tasks. Moreover, DistilHuBERT required little training time and data, opening the possibilities of pre-training personal and on-device SSL models for speech.
The original model can be found under https://github.com/s3prl/s3prl/tree/master/s3prl/upstream/distiller .
See this blog for more information on how to fine-tune the model. Note that the class Wav2Vec2ForCTC has to be replaced by HubertForCTC.
