stdbug
/

whisper-tiny-ba

@@ -1,101 +1,41 @@
 ---
-library_name: transformers
-license: apache-2.0
 base_model: openai/whisper-tiny
-tags:
-- generated_from_trainer
 datasets:
 - common_voice_17_0
-metrics:
-- wer
 model-index:
-- name: whisper-tiny-ba
   results:
   - task:
-      name: Automatic Speech Recognition
       type: automatic-speech-recognition
     dataset:
-      name: common_voice_17_0
-      type: common_voice_17_0
-      config: ba
-      split: test
-      args: ba
     metrics:
-    - name: Wer
-      type: wer
-      value: 102.54367732149878
 ---
-<!-- This model card has been generated automatically according to the information the Trainer had access to. You
-should probably proofread and complete it, then remove this comment. -->
-# whisper-tiny-ba
-This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the common_voice_17_0 dataset.
-It achieves the following results on the evaluation set:
-- Loss: 1.4410
-- Model Preparation Time: 0.0025
-- Wer Ortho: 103.0488
-- Wer: 102.5437
-- Cer Ortho: 89.2934
-- Cer: 89.2773
-## Model description
-More information needed
-## Intended uses & limitations
-More information needed
-## Training and evaluation data
-More information needed
-## Training procedure
-### Training hyperparameters
-The following hyperparameters were used during training:
-- learning_rate: 1e-05
-- train_batch_size: 8
-- eval_batch_size: 8
-- seed: 42
-- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
-- lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 50
-- training_steps: 100
-- mixed_precision_training: Native AMP
-### Training results
-| Training Loss | Epoch  | Step | Validation Loss | Model Preparation Time | Wer Ortho | Wer      | Cer Ortho | Cer      |
-|:-------------:|:------:|:----:|:---------------:|:----------------------:|:---------:|:--------:|:---------:|:--------:|
-| 5.7178        | 0.0003 | 5    | 5.8310          | 0.0025                 | 127.8009  | 150.7646 | 115.4309  | 116.2239 |
-| 5.6294        | 0.0006 | 10   | 5.6943          | 0.0025                 | 128.1157  | 154.4284 | 116.5114  | 117.1299 |
-| 5.2642        | 0.0009 | 15   | 4.9723          | 0.0025                 | 129.6523  | 168.3434 | 118.4053  | 118.6751 |
-| 4.5346        | 0.0012 | 20   | 4.1850          | 0.0025                 | 132.4879  | 180.5393 | 116.1877  | 115.6355 |
-| 3.8708        | 0.0015 | 25   | 3.7297          | 0.0025                 | 126.0743  | 147.5232 | 93.8656   | 92.4698  |
-| 3.4404        | 0.0018 | 30   | 3.3208          | 0.0025                 | 124.7111  | 127.3794 | 81.5634   | 80.3320  |
-| 3.1322        | 0.0021 | 35   | 2.9759          | 0.0025                 | 134.3791  | 135.3048 | 96.6351   | 96.8721  |
-| 2.7565        | 0.0024 | 40   | 2.6495          | 0.0025                 | 177.1889  | 177.7934 | 146.4243  | 149.6196 |
-| 2.3623        | 0.0027 | 45   | 2.3343          | 0.0025                 | 137.9902  | 138.3922 | 135.2813  | 138.0122 |
-| 2.1321        | 0.0030 | 50   | 2.1053          | 0.0025                 | 107.1350  | 107.1311 | 132.8140  | 135.4967 |
-| 2.0069        | 0.0033 | 55   | 1.9608          | 0.0025                 | 106.1485  | 108.3771 | 105.5964  | 106.4587 |
-| 1.9089        | 0.0036 | 60   | 1.8343          | 0.0025                 | 117.6690  | 123.4953 | 135.8821  | 138.0831 |
-| 1.7768        | 0.0039 | 65   | 1.7256          | 0.0025                 | 118.6013  | 130.0893 | 86.3786   | 85.1530  |
-| 1.6978        | 0.0042 | 70   | 1.6329          | 0.0025                 | 104.4993  | 109.0602 | 108.1573  | 109.0689 |
-| 1.5043        | 0.0045 | 75   | 1.5615          | 0.0025                 | 102.4446  | 102.6448 | 112.5556  | 113.9368 |
-| 1.446         | 0.0048 | 80   | 1.5109          | 0.0025                 | 103.3548  | 102.8291 | 76.8644   | 76.1611  |
-| 1.4638        | 0.0051 | 85   | 1.4736          | 0.0025                 | 103.9204  | 103.1189 | 71.3541   | 70.3195  |
-| 1.4357        | 0.0054 | 90   | 1.4410          | 0.0025                 | 103.0488  | 102.5437 | 89.2934   | 89.2773  |
-| 1.4126        | 0.0057 | 95   | 1.4188          | 0.0025                 | 102.9980  | 102.9380 | 91.5422   | 91.7081  |
-| 1.3226        | 0.0060 | 100  | 1.4030          | 0.0025                 | 104.3247  | 103.9414 | 84.2527   | 83.9457  |
-### Framework versions
-- Transformers 4.48.1
-- Pytorch 2.5.1+cu124
-- Datasets 3.2.0
-- Tokenizers 0.21.0

 ---
 base_model: openai/whisper-tiny
 datasets:
 - common_voice_17_0
+language: ba
+library_name: transformers
+license: apache-2.0
 model-index:
+- name: Finetuned openai/whisper-tiny on Bashkir
   results:
   - task:
       type: automatic-speech-recognition
+      name: Speech-to-Text
     dataset:
+      name: Common Voice (Bashkir)
+      type: common_voice
     metrics:
+    - type: wer
+      value: 102.544
 ---
+# Finetuned openai/whisper-tiny on 133675 Bashkir training audio samples from mozilla-foundation/common_voice_17_0.
+This model was created from the Mozilla.ai Blueprint:
+[speech-to-text-finetune](https://github.com/mozilla-ai/speech-to-text-finetune).
+## Evaluation results on 14513 audio samples of Bashkir:
+### Baseline model (before finetuning) on Bashkir
+- Word Error Rate (Normalized): 150.765
+- Word Error Rate (Orthographic): 127.801
+- Character Error Rate (Normalized): 116.224
+- Character Error Rate (Orthographic): 115.431
+- Loss: 5.831
+### Finetuned model (after finetuning) on Bashkir
+- Word Error Rate (Normalized): 102.544
+- Word Error Rate (Orthographic): 103.049
+- Character Error Rate (Normalized): 89.277
+- Character Error Rate (Orthographic): 89.293
+- Loss: 1.441