lucidrains/musiclm-pytorch

Health v2 · maintenance only

Adoption not part of the score

3293 stars · 264 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A PyTorch library implementing MusicLM, Google's text-to-music generation model, by combining text-conditioned AudioLM with MuLan, a text-audio contrastive embedding model. It provides the building blocks (MuLaN, AudioSpectrogramTransformer, TextTransformer, MuLaNEmbedQuantizer) needed to train and condition music generation from text descriptions.

Use cases

• generate music from text descriptions in pytorch
• train a text-audio contrastive model like MuLan
• replicate Google's MusicLM model
• embed audio and text into a joint embedding space
• condition AudioLM on text for music synthesis

When to choose

• you want to experiment with or replicate MusicLM in PyTorch
• you need a MuLan-style text-audio contrastive model implementation
• you're doing research on text-to-music generation

When to avoid

• you need a ready-to-use pretrained music generator out of the box
• you want a production music synthesis service rather than research code
• you're unwilling to train the MuLan model yourself on large audio-text datasets

Facets

library · maturity experimental

machine-learning deep-learning audio-processing llm-training deep-learning machine-learning artificial-intelligence python text-to-music music-generation contrastive-learning transformers pytorch research-replication audio

Member repositories

For agents

markdown · JSON · MCP: product_card(name="lucidrains/musiclm-pytorch")