Wavlm Base

🚀 WavLM-Base

The base model pretrained on 16kHz sampled speech audio, suitable for various speech processing tasks.

🚀 Quick Start

The base model is pretrained on 16kHz sampled speech audio. When using the model, ensure that your speech input is also sampled at 16kHz.

Note: This model does not have a tokenizer as it was pretrained on audio alone. For speech recognition, a tokenizer should be created and the model should be fine - tuned on labeled text data. Check out this blog for a more detailed explanation of how to fine - tune the model.

✨ Features

• Universal Representation: WavLM is designed to solve full - stack downstream speech tasks, aiming to learn universal representations for all speech tasks.
• Improved Structure: It equips the Transformer structure with gated relative position bias to enhance its recognition task capabilities.
• Speaker Discrimination: An utterance mixing training strategy is proposed for better speaker discrimination.
• Large - Scale Training: The training dataset is scaled up from 60k hours to 94k hours.

📦 Installation

No specific installation steps are provided in the original document, so this section is skipped.

💻 Usage Examples

Basic Usage

This is an English pre - trained speech model. It needs to be fine - tuned on a downstream task like speech recognition or audio classification before it can be used in inference. The model was pre - trained in English and performs well only in English. It has shown good performance on the SUPERB benchmark.

Note: The model was pre - trained on phonemes rather than characters. One should convert the input text to a sequence of phonemes before fine - tuning.

Advanced Usage

Speech Recognition

To fine - tune the model for speech recognition, see [the official speech recognition example](https://github.com/huggingface/transformers/tree/master/examples/pytorch/speech - recognition).

Speech Classification

To fine - tune the model for speech classification, see [the official audio classification example](https://github.com/huggingface/transformers/tree/master/examples/pytorch/audio - classification).

Speaker Verification

TODO

Speaker Diarization

TODO

📚 Documentation

Model Information

Paper Information

• Title: WavLM: Large - Scale Self - Supervised Pre - Training for Full Stack Speech Processing
• Authors: Sanyuan Chen, Chengyi Wang, Zhengyang Chen, Yu Wu, Shujie Liu, Zhuo Chen, Jinyu Li, Naoyuki Kanda, Takuya Yoshioka, Xiong Xiao, Jian Wu, Long Zhou, Shuo Ren, Yanmin Qian, Yao Qian, Jian Wu, Michael Zeng, Furu Wei
• Abstract: Self - supervised learning (SSL) achieves great success in speech recognition, while limited exploration has been attempted for other speech processing tasks. As speech signal contains multi - faceted information including speaker identity, paralinguistics, spoken content, etc., learning universal representations for all speech tasks is challenging. In this paper, we propose a new pre - trained model, WavLM, to solve full - stack downstream speech tasks. WavLM is built based on the HuBERT framework, with an emphasis on both spoken content modeling and speaker identity preservation. We first equip the Transformer structure with gated relative position bias to improve its capability on recognition tasks. For better speaker discrimination, we propose an utterance mixing training strategy, where additional overlapped utterances are created unsupervisely and incorporated during model training. Lastly, we scale up the training dataset from 60k hours to 94k hours. WavLM Large achieves state - of - the - art performance on the SUPERB benchmark, and brings significant improvements for various speech processing tasks on their representative benchmarks.

🔧 Technical Details

The original model can be found under https://github.com/microsoft/unilm/tree/master/wavlm.

📄 License

The official license can be found here

👥 Contribution

The model was contributed by cywang and patrickvonplaten.