skip to content
gigarouter gigarouter
models / embeddings · coming soon

WavLM Large

microsoft/wavlm-large

published Mar 2022 · updated Feb 2022

WavLM Large is a speech embedding model that generates deep speech representations for a variety of downstream tasks.

status
coming soon
API providers
0
downloads / mo
1.4M

specs

TaskSpeech Embedding
ArchitectureHuBERT-based Transformer with gated relative position bias and utterance mixing
Pre-training Data94,000 hours (Libri-Light 60k, GigaSpeech 10k, VoxPopuli 24k)
LicenseCC BY-SA 3.0

about this model

WavLM-Large is a self-supervised speech embedding model that produces dense vector representations from 16 kHz audio input. It is designed to capture both spoken content and speaker identity, making it suitable for a wide range of downstream tasks such as speaker verification, speech recognition, and speech separation.

Pre-training Data

The model was pre-trained on 94,000 hours of English speech, composed of 60,000 hours from Libri-Light, 10,000 hours from GigaSpeech, and 24,000 hours from VoxPopuli. This large and diverse corpus enables robust generalization across speaking styles, recording conditions, and speaker characteristics.

Architecture Highlights

Built on the HuBERT framework, WavLM incorporates a gated relative position bias in its Transformer structure to improve recognition tasks. An utterance mixing training strategy is used to enhance speaker discrimination by creating overlapping utterances during pre-training. The large variant contains approximately 300 million parameters.

Benchmark Performance

WavLM-Large achieves state-of-the-art results on the SUPERB benchmark and strong results on specialized tasks. Key results from the original repository include:

Speaker Verification on VoxCeleb1 (EER, lower is better)

DatasetEER (%)
Vox1-O0.330
Vox1-E0.477
Vox1-H0.984

Speech Separation on LibriCSS (WER, lower is better)

  • Clean (0S): 4.3%
  • Clean (0L): 4.2%
  • Overlap 10% (OV10): 5.0%
  • Overlap 20% (OV20): 6.3%
  • Overlap 30% (OV30): 8.2%
  • Overlap 40% (OV40): 8.8%

These results outperform prior systems including HuBERT large, Wav2Vec2.0 XLSR, and Conformer.

Illustration of WavLM pre-training with utterance mixing strategy

License

The model is released under the Creative Commons Attribution-ShareAlike 3.0 Unported (CC BY-SA 3.0) license.

best for

FAQ

What audio input format does WavLM Large expect?

The model expects mono audio sampled at 16 kHz. Input should be provided as raw audio waveforms.

Can I use WavLM Large as an embedding model without fine-tuning?

Yes, you can extract frame-level speech representations from the pre-trained model and use them as embeddings for downstream tasks.

What are the license terms for using WavLM Large?

The model is licensed under Creative Commons Attribution-ShareAlike 3.0 Unported (CC BY-SA 3.0).

How does WavLM Large perform on speaker verification benchmarks?

On VoxCeleb1, it achieves 0.33% equal error rate (EER) on the Vox1-O test set when fine-tuned.

How do I call WavLM Large via the gigarouter API?

Use the OpenAI-compatible endpoint with your gigarouter API key. Send a request with the audio file or base64-encoded audio.

not yet live

We're benchmarking and onboarding WavLM Large as a hosted, OpenAI-compatible API. Sign in for free credit and be ready when it lands, or tell us you want it and we'll prioritize it.

related embeddings models

compare all →