mazesmazes
/

tiny-audio

Automatic Speech Recognition

feature-extraction

speech-recognition

Model card Files Files and versions

mazesmazes commited on 7 days ago

Commit

0366e08

·

verified ·

1 Parent(s): d8c53de

Training in progress, step 1000

Files changed (3) hide show

config.json +2 -2
model.safetensors +2 -2
training_args.bin +2 -2

config.json CHANGED Viewed

@@ -180,7 +180,7 @@
   "projector_init_std": 0.02,
   "projector_num_layers": 2,
   "projector_pool_stride": 4,
-  "projector_type": "linear",
   "qformer_hidden_size": null,
   "qformer_intermediate_size": null,
   "qformer_num_heads": 16,
@@ -254,7 +254,7 @@
   "text_model_id": "Qwen/Qwen3-0.6B",
   "transformers_version": "5.0.0.dev0",
   "use_cache": false,
-  "use_specaugment": true,
   "user_prompt": "Please transcribe this English audio into text: <audio>",
   "vocab_size": 151670
 }

   "projector_init_std": 0.02,
   "projector_num_layers": 2,
   "projector_pool_stride": 4,
+  "projector_type": "mlp",
   "qformer_hidden_size": null,
   "qformer_intermediate_size": null,
   "qformer_num_heads": 16,
   "text_model_id": "Qwen/Qwen3-0.6B",
   "transformers_version": "5.0.0.dev0",
   "use_cache": false,
+  "use_specaugment": false,
   "user_prompt": "Please transcribe this English audio into text: <audio>",
   "vocab_size": 151670
 }

model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a82f0e82a684298cc550fefb3a4968fa974cf79a2568d441bcf3e4b350a76f23
-size 2623704

 version https://git-lfs.github.com/spec/v1
+oid sha256:bc90ebd7f1e678a85a0a5e78f37c8a70357d296b72ad3a9a457d9359cc2ae7d4
+size 25172384

training_args.bin CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1207623a291f4a86f41326955931e7f95611cf08e5ed2ba9c8eeebca5dc00b18
-size 5265

 version https://git-lfs.github.com/spec/v1
+oid sha256:66e4fd5269ddd0a85cab49962bc693e29b8e66f64f6a5ac11347bafc2311b7e3
+size 5201