add model

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "wav2vec2-xls-r-1b-5gram-german",
+  "_name_or_path": "wav2vec2-xls-r-1b-german",
   "activation_dropout": 0.0,
   "adapter_kernel_size": 3,
   "adapter_stride": 2,
@@ -11,9 +11,9 @@
   "attention_dropout": 0.0,
   "bos_token_id": 1,
   "classifier_proj_size": 256,
-  "codevector_dim": 256,
+  "codevector_dim": 1024,
   "contrastive_logits_temperature": 0.1,
-  "conv_bias": false,
+  "conv_bias": true,
   "conv_dim": [
     512,
     512,
@@ -44,50 +44,40 @@
   "ctc_loss_reduction": "mean",
   "ctc_zero_infinity": false,
   "diversity_loss_weight": 0.1,
-  "do_stable_layer_norm": false,
+  "do_stable_layer_norm": true,
   "eos_token_id": 2,
   "feat_extract_activation": "gelu",
-  "feat_extract_norm": "group",
+  "feat_extract_dropout": 0.0,
+  "feat_extract_norm": "layer",
   "feat_proj_dropout": 0.0,
   "feat_quantizer_dropout": 0.0,
   "final_dropout": 0.0,
-  "freeze_feat_extract_train": true,
   "hidden_act": "gelu",
   "hidden_dropout": 0.0,
-  "hidden_size": 768,
+  "hidden_size": 1280,
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
+  "intermediate_size": 5120,
   "layer_norm_eps": 1e-05,
   "layerdrop": 0.0,
-  "mask_channel_length": 10,
-  "mask_channel_min_space": 1,
-  "mask_channel_other": 0.0,
-  "mask_channel_prob": 0.0,
-  "mask_channel_selection": "static",
   "mask_feature_length": 10,
   "mask_feature_min_masks": 0,
   "mask_feature_prob": 0.0,
   "mask_time_length": 10,
   "mask_time_min_masks": 2,
-  "mask_time_min_space": 1,
-  "mask_time_other": 0.0,
   "mask_time_prob": 0.1,
-  "mask_time_selection": "static",
   "model_type": "wav2vec2",
-  "no_mask_channel_overlap": false,
-  "no_mask_time_overlap": false,
   "num_adapter_layers": 3,
-  "num_attention_heads": 12,
+  "num_attention_heads": 16,
   "num_codevector_groups": 2,
   "num_codevectors_per_group": 320,
   "num_conv_pos_embedding_groups": 16,
   "num_conv_pos_embeddings": 128,
   "num_feat_extract_layers": 7,
-  "num_hidden_layers": 12,
+  "num_hidden_layers": 48,
   "num_negatives": 100,
-  "output_hidden_size": 768,
+  "output_hidden_size": 1280,
   "pad_token_id": 64,
-  "proj_codevector_dim": 256,
+  "proj_codevector_dim": 1024,
   "tdnn_dilation": [
     1,
     2,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bf8b03e848773744ee0f27223a98818b03c2e23e3049e9982c7dac3e46437d13
-size 377761559
+oid sha256:ce86c94dc1434b1dfd65b935bb0ba381f45c7a7c95ccaf7d9940776531ded836
+size 3850597041