Project AURIONAI Use-Cases Research & Innovation
Microsoft AI
speech recognitionfamily

WavLM

Self-supervised speech model trained for full-stack speech processing, including speaker, paralinguistic and content tasks.

Microsoft AIMicrosoft AIUSA
Diagram of the WavLM model with CNN encoders and a transformer trained by masked prediction.
Levelfamily
Classspeech recognition
Parametersnot in the notes
Contextnot in the notes
Released26 Oct 2021
Opennessopen weights permissive
Runs onnot in the notes
Statusactive

Lineage

Microsoft AIwavlm-base-plus belongs to this familyspeech recognition26 Oct 2021
Microsoft AIwavlm-large belongs to this familyspeech recognition26 Oct 2021
Where this comes from
Parent and child relations from model cards, Hugging Face base_model fields and the genealogy tables of the research notes. Confidence and evidence are the notes' own.

R&D Topics

Self-supervised speech representation learningexemplar

Identity

Developer
Microsoft AI
Family
not in the notes
Released
26 Oct 2021 sourcesannouncement page
Weights published
–
Record checked
12 Oct 2026
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.

Architecture and Size

Architecture
dense transformer self-supervised masked speech prediction with denoising
Total parameters
– sources
Active parameters
–
Checkpoint parameters
–
Note
Base, Base Plus and Large checkpoints.
Context length
not in the notes
Input
audio
Output
not in the notes
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).

Openness and Access

Openness
open weights permissive
Licence
MIT License sources
Access routes
open weights download
Where this comes from
The licence file governs over registry tags and aggregator labels.

Deployment

Runs on
not in the notes
Footprint
Licence taken from the microsoft/unilm repository LICENSE (MIT); the Hugging Face cards carry no licence tag.
Runtimes
not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.

Indonesia

Indonesian support
not stated
Regional languages
not in the notes
Jakarta hosting
unknown
Languages
not in the notes
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.

Benchmark Results

The notes record no benchmark result for this model.

Industry Use Cases

No industry use case links this model.

Sources

  1. arxiv.org/abs/2110.13900 https://arxiv.org/abs/2110.13900
  2. github.com/microsoft/unilm/tree/master/wavlm https://github.com/microsoft/unilm/tree/master/wavlm
  3. huggingface.co/microsoft/wavlm-large https://huggingface.co/microsoft/wavlm-large

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "WavLM", https://aurion.semai.id/models/microsoft-wavlm/, records of October 2026.