speecht5_finetuned_wolof_text_speech_2

This model is a fine-tuned version of microsoft/speecht5_tts on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3084

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: 0.0001
• train_batch_size: 4
• eval_batch_size: 2
• seed: 42
• 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: 100
• training_steps: 15000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.3484	10.6809	1000	0.3243
0.3243	21.3618	2000	0.3130
0.3181	32.0427	3000	0.3127
0.3053	42.7236	4000	0.3099
0.2973	53.4045	5000	0.3069
0.2971	64.0854	6000	0.3091
0.2913	74.7664	7000	0.3058
0.2851	85.4473	8000	0.3077
0.284	96.1282	9000	0.3073
0.2809	106.8091	10000	0.3066
0.2796	117.4900	11000	0.3083
0.2747	128.1709	12000	0.3068
0.2767	138.8518	13000	0.3083
0.2738	149.5327	14000	0.3075
0.2748	160.2136	15000	0.3084

Framework versions

• Transformers 4.45.1
• Pytorch 2.4.0
• Datasets 3.0.1
• Tokenizers 0.20.0