Training in progress, step 50

README.md

@@ -1,35 +1,83 @@
 ---
-base_model: openai/whisper-small
-datasets:
-- mozilla-foundation/common_voice_17_0
-language: sw
 library_name: transformers
 license: apache-2.0
+base_model: openai/whisper-small
+tags:
+- generated_from_trainer
+datasets:
+- common_voice_17_0
+metrics:
+- wer
 model-index:
-- name: Finetuned openai/whisper-small on Swahili
+- name: ASR-Swahili-Small
   results:
   - task:
+      name: Automatic Speech Recognition
       type: automatic-speech-recognition
-      name: Speech-to-Text
     dataset:
-      name: Common Voice (Swahili)
-      type: common_voice
+      name: common_voice_17_0
+      type: common_voice_17_0
+      config: sw
+      split: test
+      args: sw
     metrics:
-    - type: wer
-      value: 53.403
+    - name: Wer
+      type: wer
+      value: 53.40302063717767
 ---
 
-# Finetuned openai/whisper-small on 20000 Swahili training audio samples from mozilla-foundation/common_voice_17_0.
+<!-- 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. -->
+
+# ASR-Swahili-Small
+
+This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the common_voice_17_0 dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.8496
+- Model Preparation Time: 0.003
+- Wer: 53.4030
+
+## 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: 32
+- eval_batch_size: 16
+- seed: 42
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 128
+- 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
+- num_epochs: 1
+- mixed_precision_training: Native AMP
+
+### Training results
 
-This model was created from the Mozilla.ai Blueprint:
-[speech-to-text-finetune](https://github.com/mozilla-ai/speech-to-text-finetune).
+| Training Loss | Epoch | Step | Validation Loss | Model Preparation Time | Wer     |
+|:-------------:|:-----:|:----:|:---------------:|:----------------------:|:-------:|
+| 1.5828        | 0.32  | 50   | 1.1291          | 0.003                  | 63.7351 |
+| 0.8342        | 0.64  | 100  | 0.8994          | 0.003                  | 56.2661 |
+| 0.7015        | 0.96  | 150  | 0.8496          | 0.003                  | 53.4030 |
 
-## Evaluation results on 12253 audio samples of Swahili:
 
-### Baseline model (before finetuning) on Swahili
-- Word Error Rate: 133.795
-- Loss: 2.459
+### Framework versions
 
-### Finetuned model (after finetuning) on Swahili
-- Word Error Rate: 53.403
-- Loss: 0.85
+- Transformers 4.49.0
+- Pytorch 2.6.0+cu124
+- Datasets 3.3.1
+- Tokenizers 0.21.0

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:abb8cd0ad7f71fea01b8a372781d7f31adb335f72cf07afb5f46b7f21e431291
+oid sha256:0f6dd8d26a8176f8e1db1e0460e84eccb4a75bf4a4f20bb8d835e334cd491c2b
 size 966995080

runs/Feb27_15-10-06_ai4d-Lambda-Vector/events.out.tfevents.1740660521.ai4d-Lambda-Vector.2836067.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5c4add46f8e9cfaa65bc92bb97d2ed97b95156b0ddcbe49ac984d026306aacb
+size 7765

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48f921cced55cf902a140bcbd2244853c66a2b1fd78f76f701653cbf83d66c20
+oid sha256:2ec319de276aa407a885fa77816c52eeff7a7b42e57dde68a6af9e7616ad278e
 size 5496