whisper-tiny-cv-full-synthetic-pt

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

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 5e-05
train_batch_size: 256
eval_batch_size: 8
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 10

Training Loss	Epoch	Step	Validation Loss
0.6982	0.2907	50	0.6187
0.5012	0.5814	100	0.5461
0.448	0.8721	150	0.5133
0.3321	1.1628	200	0.4977
0.3198	1.4535	250	0.4814
0.3069	1.7442	300	0.4700
0.2712	2.0349	350	0.4582
0.1954	2.3256	400	0.4599
0.2032	2.6163	450	0.4585
0.2011	2.9070	500	0.4517
0.1334	3.1977	550	0.4576
0.1309	3.4884	600	0.4596
0.13	3.7791	650	0.4613
0.1146	4.0698	700	0.4667
0.0906	4.3605	750	0.4702
0.0918	4.6512	800	0.4754
0.0955	4.9419	850	0.4741
0.0599	5.2326	900	0.4841
0.0647	5.5233	950	0.4877
0.0611	5.8140	1000	0.4953
0.0505	6.1047	1050	0.5025
0.042	6.3953	1100	0.5023
0.0448	6.6860	1150	0.5082
0.0452	6.9767	1200	0.5040
0.0294	7.2674	1250	0.5132
0.032	7.5581	1300	0.5164
0.0309	7.8488	1350	0.5183
0.0239	8.1395	1400	0.5241
0.0232	8.4302	1450	0.5250
0.0244	8.7209	1500	0.5256
0.024	9.0116	1550	0.5278
0.0207	9.3023	1600	0.5291
0.0217	9.5930	1650	0.5303
0.0203	9.8837	1700	0.5305