Loading audiolm_pytorch_demo_laion.py +4 −4 Original line number Diff line number Diff line Loading @@ -70,7 +70,7 @@ make_placeholder_dataset() ####### # soundstream = AudioLMSoundStream( # codec = AudioLMSoundStream( # codebook_size = 1024, # rq_num_quantizers = 8, # attn_window_size = 128, # local attention receptive field at bottleneck Loading Loading @@ -133,7 +133,7 @@ coarse_transformer = CoarseTransformer( coarse_trainer = CoarseTransformerTrainer( transformer = coarse_transformer, soundstream = encodec, codec = encodec, wav2vec = wav2vec, folder = dataset_folder, batch_size = 1, Loading @@ -160,7 +160,7 @@ fine_transformer = FineTransformer( fine_trainer = FineTransformerTrainer( transformer = fine_transformer, soundstream = encodec, codec = encodec, folder = dataset_folder, batch_size = 1, data_max_length = 320 * 32, Loading @@ -179,7 +179,7 @@ fine_trainer.train() audiolm = AudioLM( wav2vec = wav2vec, soundstream = encodec, codec = encodec, semantic_transformer = semantic_transformer, coarse_transformer = coarse_transformer, fine_transformer = fine_transformer Loading Loading
audiolm_pytorch_demo_laion.py +4 −4 Original line number Diff line number Diff line Loading @@ -70,7 +70,7 @@ make_placeholder_dataset() ####### # soundstream = AudioLMSoundStream( # codec = AudioLMSoundStream( # codebook_size = 1024, # rq_num_quantizers = 8, # attn_window_size = 128, # local attention receptive field at bottleneck Loading Loading @@ -133,7 +133,7 @@ coarse_transformer = CoarseTransformer( coarse_trainer = CoarseTransformerTrainer( transformer = coarse_transformer, soundstream = encodec, codec = encodec, wav2vec = wav2vec, folder = dataset_folder, batch_size = 1, Loading @@ -160,7 +160,7 @@ fine_transformer = FineTransformer( fine_trainer = FineTransformerTrainer( transformer = fine_transformer, soundstream = encodec, codec = encodec, folder = dataset_folder, batch_size = 1, data_max_length = 320 * 32, Loading @@ -179,7 +179,7 @@ fine_trainer.train() audiolm = AudioLM( wav2vec = wav2vec, soundstream = encodec, codec = encodec, semantic_transformer = semantic_transformer, coarse_transformer = coarse_transformer, fine_transformer = fine_transformer Loading