Add model weights

model_index.json

@@ -0,0 +1,28 @@
+{
+  "_class_name": "AudioLDMPipeline",
+  "_diffusers_version": "0.13.0.dev0",
+  "scheduler": [
+    "diffusers",
+    "DDIMScheduler"
+  ],
+  "text_encoder": [
+    "transformers",
+    "ClapTextModelWithProjection"
+  ],
+  "tokenizer": [
+    "transformers",
+    "RobertaTokenizerFast"
+  ],
+  "unet": [
+    "diffusers",
+    "UNet2DModel"
+  ],
+  "vae": [
+    "diffusers",
+    "AutoencoderKL"
+  ],
+  "vocoder": [
+    "transformers",
+    "SpeechT5HifiGan"
+  ]
+}

scheduler/scheduler_config.json

@@ -0,0 +1,13 @@
+{
+  "_class_name": "DDIMScheduler",
+  "_diffusers_version": "0.13.0.dev0",
+  "beta_end": 0.0195,
+  "beta_schedule": "scaled_linear",
+  "beta_start": 0.0015,
+  "clip_sample": false,
+  "num_train_timesteps": 1000,
+  "prediction_type": "epsilon",
+  "set_alpha_to_one": false,
+  "steps_offset": 1,
+  "trained_betas": null
+}

text_encoder/config.json

@@ -0,0 +1,32 @@
+{
+  "architectures": [
+    "ClapTextModelWithProjection"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "fusion_hidden_size": 768,
+  "fusion_num_hidden_layers": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_factor": 1.0,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 514,
+  "model_type": "clap_text_model",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "projection_dim": 512,
+  "projection_hidden_act": "relu",
+  "projection_hidden_size": 768,
+  "torch_dtype": "float32",
+  "transformers_version": "4.27.0.dev0",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50265
+}

text_encoder/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0292e65e99024f2d21f0a59b718f0e4914546e4287e2dabdd2d7e4a95c169f7b
+size 501284353

tokenizer/special_tokens_map.json

@@ -0,0 +1,15 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

tokenizer/tokenizer_config.json

@@ -0,0 +1,16 @@
+{
+  "add_prefix_space": false,
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "errors": "replace",
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "processor_class": "ClapProcessor",
+  "sep_token": "</s>",
+  "special_tokens_map_file": null,
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
+}

unet/config.json

@@ -0,0 +1,43 @@
+{
+  "_class_name": "UNet2DModel",
+  "_diffusers_version": "0.13.0.dev0",
+  "act_fn": "silu",
+  "add_attention": true,
+  "attention_head_dim": 8,
+  "block_out_channels": [
+    128,
+    256,
+    384,
+    640
+  ],
+  "center_input_sample": false,
+  "class_embed_type": null,
+  "cross_attention_dim": true,
+  "down_block_types": [
+    "DownBlock2D",
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D",
+    "CrossAttnDownBlock2D"
+  ],
+  "downsample_padding": 1,
+  "extra_film_condition_dim": 512,
+  "extra_film_use_concat": true,
+  "flip_sin_to_cos": true,
+  "freq_shift": 0,
+  "in_channels": 8,
+  "layers_per_block": 2,
+  "mid_block_scale_factor": 1,
+  "norm_eps": 1e-05,
+  "norm_num_groups": 32,
+  "num_class_embeds": null,
+  "out_channels": 8,
+  "resnet_time_scale_shift": "default",
+  "sample_size": 128,
+  "time_embedding_type": "positional",
+  "up_block_types": [
+    "CrossAttnUpBlock2D",
+    "CrossAttnUpBlock2D",
+    "CrossAttnUpBlock2D",
+    "UpBlock2D"
+  ]
+}

unet/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51f71fa4e1233882e2a5cd2cd7ed4cf64d11d59a6c8c0b7a09358c58f36ef9f3
+size 740429981

vae/config.json

@@ -0,0 +1,27 @@
+{
+  "_class_name": "AutoencoderKL",
+  "_diffusers_version": "0.13.0.dev0",
+  "act_fn": "silu",
+  "block_out_channels": [
+    128,
+    256,
+    512
+  ],
+  "down_block_types": [
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D"
+  ],
+  "in_channels": 1,
+  "latent_channels": 8,
+  "layers_per_block": 2,
+  "norm_num_groups": 32,
+  "out_channels": 1,
+  "sample_size": 512,
+  "scaling_factor": 0.9227914214134216,
+  "up_block_types": [
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D"
+  ]
+}

vae/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3fc8ccecb1849c8a23cd4f9dd959eb7aaa203cc010386288418dbc551cdaaf7
+size 221586505

vocoder/config.json

@@ -0,0 +1,50 @@
+{
+  "architectures": [
+    "SpeechT5HifiGan"
+  ],
+  "initializer_range": 0.01,
+  "leaky_relu_slope": 0.1,
+  "model_in_dim": 64,
+  "model_type": "hifigan",
+  "normalize_before": false,
+  "resblock_dilation_sizes": [
+    [
+      1,
+      3,
+      5
+    ],
+    [
+      1,
+      3,
+      5
+    ],
+    [
+      1,
+      3,
+      5
+    ]
+  ],
+  "resblock_kernel_sizes": [
+    3,
+    7,
+    11
+  ],
+  "sampling_rate": 16000,
+  "torch_dtype": "float32",
+  "transformers_version": "4.27.0.dev0",
+  "upsample_initial_channel": 1024,
+  "upsample_kernel_sizes": [
+    16,
+    16,
+    8,
+    4,
+    4
+  ],
+  "upsample_rates": [
+    5,
+    4,
+    2,
+    2,
+    2
+  ]
+}

vocoder/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9fbefc2b31c85d1dabe98e53d09ac88039af411162a7e641040a9c2b5f62364
+size 221120349