zhvng / open-musiclm

Implementation of MusicLM, a text to music model published by Google Research, with a few modifications.

https://arxiv.org/abs/2301.11325

zhvng/open-musiclm Issues

empty db, curious errors, empty output, long gen time, outputs noise
Updated 7 months ago
can't train the clap
Updated 7 months ago3
What data did you use for training
Updated 7 months ago
Ran the reasoning for hours without stopping
Closed 10 months ago1
RuntimeError: Error(s) in loading state_dict for TokenConditionedTransformer
Updated 8 months ago2
Question regarding dataset
Updated 10 months ago
The question about training CLAP RVQ
Updated a year ago4
train loss of semantic stage has problem
Closed a year ago3
Can you share the CLAP RVQ checkpoints?
Closed a year ago
Invalid float value(s)
Closed a year ago
how can i train the model or exist pre trained to use
Updated a year ago
conda env create -f environment.yaml fails in fresh conda install
Closed a year ago1
Only generate a melody and not a song with lyrics?
Updated a year ago
Discord invite link expired
Closed a year ago3
training data
Updated a year ago
Is there too little data for clustering？
Closed a year ago
Models workings, but audio quality is not good.
Closed a year ago12
empty data_processor
Closed a year ago2
Pretrained models
Closed a year ago1
loading clap checkpoint
Closed a year ago
ClapRVQTrainer does not have log_with
Closed a year ago1
Hubert args normalization
Closed a year ago2
Installation and Basic Usage Guide
Closed a year ago2
Error when trying to train CLAP
Closed a year ago5