audiocraft
audiocraft copied to clipboard
facebookresearch
Audiocraft is a library for audio processing and generation with deep learning. It features the state-of-the-art EnCodec audio compressor / tokenizer, along with MusicGen, a simple and controllable mu...
The problem: With long generations, volume tends to "overflow", which results in a bad quality. Possible solution: During generation, normalize volume of previously generated steps - do not allow to...
I tried 8 times waiting at least 20min, tried small and medium 5, 10, 60 and 120 sec and the loading always gets stuck Windows 10
I notice this model design is very similar to Bark. Just more efficient. Once training code is released. If pre-trained with a corpus of speech pairs, wouldn't this make a...
input: [so cute~.mp3.zip](https://github.com/facebookresearch/audiocraft/files/11876722/so.cute.mp3.zip) 2010s rock song with electric guitar and heavy drums output: [tmpflwkfvuf.mp4.zip](https://github.com/facebookresearch/audiocraft/files/11876733/tmpflwkfvuf.mp4.zip) Any suggestions regarding usage?
I am a newbie who just started to study the direction of music generation, and I have a lot of confusion about this, so I hope we can communicate with...
not sure if there is a more proper way but found that by adding to generate(), lm() and utils functions the possibility to pass a torch.generator() as argument lets me...
Hello. I tried every different setting with MusicGen, but i still got Black windows when i generated everything like this.   Am I missed something on installation?...
Unsure what this error is about. Is this to do with a HuggingFace token or a model called t5-base? File "C:\Users\ABC\AppData\Local\Programs\Python\Python310\lib\site-packages\filelock\_windows.py", line 25, in _acquire fd = os.open(self.lock_file, flags, self._context.mode)...
Would it be possible to use previous tokens to continue generation? and if so, how could it be implemented using the standard API (torch, audiocraft)?
Metadata
Owner
Metadata
Audiocraft is a library for audio processing and generation with deep learning. It features the state-of-the-art EnCodec audio compressor / tokenizer, along with MusicGen, a simple and controllable mu...