End of training

README.md

@@ -1,24 +1,23 @@
 ---
-language:
-- ha
 license: mit
 base_model: microsoft/speecht5_tts
 tags:
-- hausa
 - generated_from_trainer
 datasets:
-- mozilla-foundation/common_voice_8_0
+- common_voice_8_0
 model-index:
-- name: SpeechT5 TTS Hausa
+- name: common_voice_8_0_ha
   results: []
 ---
 
 <!-- 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. -->
 
-# SpeechT5 TTS Hausa
+# common_voice_8_0_ha
 
-This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the VoxPopuli dataset.
+This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the common_voice_8_0 dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.4741
 
 ## Model description
 
@@ -38,16 +37,26 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 1e-05
-- train_batch_size: 16
-- eval_batch_size: 8
+- train_batch_size: 4
+- eval_batch_size: 2
 - seed: 42
-- gradient_accumulation_steps: 2
+- gradient_accumulation_steps: 8
 - total_train_batch_size: 32
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 500
 - training_steps: 4000
 
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 0.5416        | 18.31 | 1000 | 0.4974          |
+| 0.505         | 36.61 | 2000 | 0.4760          |
+| 0.4898        | 54.92 | 3000 | 0.4758          |
+| 0.5004        | 73.23 | 4000 | 0.4741          |
+
+
 ### Framework versions
 
 - Transformers 4.33.0.dev0