Update tiny models for PeAudioEncoder

config.json

@@ -0,0 +1,49 @@
+{
+  "architectures": [
+    "PeAudioEncoder"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "dac_config": {
+    "codebook_dim": 32,
+    "codebook_loss_weight": 1.0,
+    "codebook_size": 512,
+    "commitment_loss_weight": 0.25,
+    "decoder_hidden_size": 16,
+    "downsampling_ratios": [
+      2,
+      4,
+      4
+    ],
+    "encoder_hidden_size": 16,
+    "hidden_size": 128,
+    "hop_length": 32,
+    "model_type": "dac",
+    "n_codebooks": 6,
+    "quantizer_dropout": 0.0,
+    "sampling_rate": 16000,
+    "upsampling_ratios": [
+      4,
+      4,
+      2
+    ]
+  },
+  "dtype": "float32",
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 32,
+  "initializer_range": 0.02,
+  "intermediate_size": 37,
+  "max_position_embeddings": 512,
+  "max_window_layers": 28,
+  "model_type": "pe_audio_encoder",
+  "num_attention_heads": 2,
+  "num_hidden_layers": 2,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-05,
+  "rope_parameters": {
+    "rope_theta": 20000,
+    "rope_type": "default"
+  },
+  "transformers_version": "5.16.0.dev0"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96bde82903f76747f702b3a8b7e83aff50796c7632413454f9918d27bae6fb2c
+size 1411832

preprocessor_config.json

@@ -0,0 +1,9 @@
+{
+  "feature_extractor_type": "PeAudioFeatureExtractor",
+  "feature_size": 1,
+  "hop_length": 1920,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "return_attention_mask": true,
+  "sampling_rate": 48000
+}

tokenizer_config.json

@@ -0,0 +1,19 @@
+{
+  "backend": "tokenizers",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "extra_special_tokens": [],
+  "is_local": true,
+  "local_files_only": false,
+  "mask_token": "[MASK]",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 512,
+  "pad_token": "[PAD]",
+  "processor_class": "PeAudioVideoProcessor",
+  "sep_token": "[SEP]",
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": "[UNK]"
+}

video_preprocessor_config.json

@@ -0,0 +1,27 @@
+{
+  "data_format": "channels_first",
+  "default_to_square": true,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "do_sample_frames": false,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "return_metadata": false,
+  "size": {
+    "height": 336,
+    "width": 336
+  },
+  "video_processor_type": "PeVideoVideoProcessor"
+}