cifope/whisper-tiny-wolof

README.md

@@ -1,6 +1,4 @@
 ---
-license: apache-2.0
-base_model: openai/whisper-tiny
 tags:
 - generated_from_trainer
 model-index:
@@ -13,7 +11,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # whisper-tiny-wolof
 
-This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the None dataset.
+This model was trained from scratch on the None dataset.
 
 ## Model description
 
@@ -33,13 +31,13 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 32
-- eval_batch_size: 4
+- train_batch_size: 48
+- eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 100
-- num_epochs: 12
+- num_epochs: 10
 - mixed_precision_training: Native AMP
 
 ### Framework versions

added_tokens.json

@@ -1603,6 +1603,7 @@
   "<|ur|>": 50290,
   "<|uz|>": 50337,
   "<|vi|>": 50278,
+  "<|wo|>": 51865,
   "<|yi|>": 50335,
   "<|yo|>": 50325,
   "<|zh|>": 50260

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "openai/whisper-tiny",
+  "_name_or_path": "/content/whisper-tiny",
   "activation_dropout": 0.0,
   "activation_function": "gelu",
   "apply_spec_augment": false,

generation_config.json

@@ -130,7 +130,7 @@
     "<|ur|>": 50290,
     "<|uz|>": 50337,
     "<|vi|>": 50278,
-    "<|wo|>": 99,
+    "<|wo|>": 100,
     "<|yi|>": 50335,
     "<|yo|>": 50325,
     "<|zh|>": 50260

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:00a0c55f325246ae86e5487571e40897fdd67ff65d0c246e0c2a6487a860faed
+oid sha256:b567e76b4f089091b1ab829df505a1138e1a7703480f1b89ad63bff2c21a8950
 size 151061672

runs/Aug12_11-57-53_5fe87a65fc63/events.out.tfevents.1723463876.5fe87a65fc63.944.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:588ba45e00eeacc41d3ef9f94c1da0387f0ca1d564cd371a22b073ff0218931a
+size 26065

special_tokens_map.json

@@ -100,6 +100,7 @@
     "<|ha|>",
     "<|ba|>",
     "<|jw|>",
+    "<|wo|>",
     "<|su|>",
     "<|translate|>",
     "<|transcribe|>",

tokenizer_config.json

@@ -12865,6 +12865,14 @@
       "rstrip": false,
       "single_word": false,
       "special": false
+    },
+    "51865": {
+      "content": "<|wo|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
     }
   },
   "additional_special_tokens": [
@@ -12968,6 +12976,7 @@
     "<|ha|>",
     "<|ba|>",
     "<|jw|>",
+    "<|wo|>",
     "<|su|>",
     "<|translate|>",
     "<|transcribe|>",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2b7267de1f8217a478ed0e065b3ddca0655779feaf1ad934e38cb001bca1a4c7
+oid sha256:78af09651e0fdaa01c18d89010a129e4895ad1b396ca79ed60cc00a2e1b74835
 size 5368