Training in progress, step 1000

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "distil-whisper/distil-large-v3",
+  "_name_or_path": "openai/whisper-tiny",
   "activation_dropout": 0.0,
   "activation_function": "gelu",
   "apply_spec_augment": false,
@@ -10,18 +10,32 @@
   "begin_suppress_tokens": null,
   "bos_token_id": 50257,
   "classifier_proj_size": 256,
-  "d_model": 1280,
-  "decoder_attention_heads": 20,
-  "decoder_ffn_dim": 5120,
+  "d_model": 384,
+  "decoder_attention_heads": 6,
+  "decoder_ffn_dim": 1536,
   "decoder_layerdrop": 0.0,
-  "decoder_layers": 2,
+  "decoder_layers": 4,
   "decoder_start_token_id": 50258,
   "dropout": 0.0,
-  "encoder_attention_heads": 20,
-  "encoder_ffn_dim": 5120,
+  "encoder_attention_heads": 6,
+  "encoder_ffn_dim": 1536,
   "encoder_layerdrop": 0.0,
-  "encoder_layers": 32,
+  "encoder_layers": 4,
   "eos_token_id": 50257,
+  "forced_decoder_ids": [
+    [
+      1,
+      50259
+    ],
+    [
+      2,
+      50359
+    ],
+    [
+      3,
+      50363
+    ]
+  ],
   "init_std": 0.02,
   "is_encoder_decoder": true,
   "mask_feature_length": 10,
@@ -35,13 +49,13 @@
   "max_target_positions": 448,
   "median_filter_width": 7,
   "model_type": "whisper",
-  "num_hidden_layers": 32,
-  "num_mel_bins": 128,
-  "pad_token_id": 50256,
+  "num_hidden_layers": 4,
+  "num_mel_bins": 80,
+  "pad_token_id": 50257,
   "scale_embedding": false,
   "torch_dtype": "float32",
   "transformers_version": "4.48.0",
   "use_cache": true,
   "use_weighted_layer_sum": false,
-  "vocab_size": 51866
+  "vocab_size": 51865
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3a69adb0c0abc466816efeba6d8f8e9cf86310f364aaa418ef5c5d8c64264ada
-size 3025686376
+oid sha256:412646c1aa1f8220c4c911258bc5cb7a8911b2a081b722ac2a4a5bbbfb148b86
+size 151061672

preprocessor_config.json

@@ -1,7 +1,7 @@
 {
   "chunk_length": 30,
   "feature_extractor_type": "WhisperFeatureExtractor",
-  "feature_size": 128,
+  "feature_size": 80,
   "hop_length": 160,
   "n_fft": 400,
   "n_samples": 480000,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:572ad06e90352905c554829c1a4e4ce912912ed596fe27cff1185bc17e9a1252
+oid sha256:02edb69986dbfaf8e07afb972b2135a2c10a9458f42fcc07ede4b26b55f9f081
 size 5432