Commit 4750ff07 authored by Leon Wu's avatar Leon Wu
Browse files

change soundstream quantizers back

parent d91ed4cb

audiolm_pytorch_demo_laion.py

+2 −2
Original line number Diff line number Diff line
@@ -70,7 +70,7 @@ make_placeholder_dataset() 


soundstream = AudioLMSoundStream(

    codebook_size = 1024,

    rq_num_quantizers = 9,

    rq_num_quantizers = 8,

    attn_window_size = 128,       # local attention receptive field at bottleneck

    attn_depth = 2                # 2 local attention transformer blocks - the soundstream folks were not experts with attention, so i took the liberty to add some. encodec went with lstms, but attention should be better

)
@@ -148,7 +148,7 @@ coarse_trainer.train() 


fine_transformer = FineTransformer(

    num_coarse_quantizers = 3,

    num_fine_quantizers = 6,

    num_fine_quantizers = 5,

    codebook_size = 1024,

    dim = 512,

    depth = 6