correct model

config.json

@@ -94,13 +94,13 @@
     "use_cache": true,
     "vocab_size": 250054
   },
-  "decoder_start_token_id": 250004,
+  "decoder_start_token_id": 2,
   "encoder": {
     "_name_or_path": "",
     "activation_dropout": 0.0,
-    "adaptor_kernel_size": 3,
-    "adaptor_stride": 2,
-    "add_adaptor": true,
+    "adapter_kernel_size": 3,
+    "adapter_stride": 2,
+    "add_adapter": true,
     "add_cross_attention": false,
     "apply_spec_augment": true,
     "architectures": [
@@ -188,7 +188,7 @@
     "min_length": 0,
     "model_type": "wav2vec2",
     "no_repeat_ngram_size": 0,
-    "num_adaptor_layers": 3,
+    "num_adapter_layers": 3,
     "num_attention_heads": 16,
     "num_beam_groups": 1,
     "num_beams": 1,
@@ -230,6 +230,8 @@
   },
   "eos_token_id": 2,
   "feature_extractor_type": "wav2vec2",
+  "forced_bos_token_id": 250004,
+  "forced_eos_token_id": 2,
   "is_encoder_decoder": true,
   "model_type": "speech-encoder-decoder",
   "pad_token_id": 1,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24a8779d697f240f846a340b9c93ade4706d2f40c126af70bf3e3f3b91a77929
+oid sha256:6b14077bddd591cdea183133ccfa800f942199affd2a87b371dd5d109eee0131
 size 5765908733