whisper_syl_noforce__0045

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

• Train Loss: 0.1117
• Train Accuracy: 0.0356
• Train Wermet: 0.0270
• Validation Loss: 0.6702
• Validation Accuracy: 0.0231
• Validation Wermet: 0.2585
• Epoch: 44

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:

• optimizer: {'name': 'AdamWeightDecay', 'learning_rate': 1e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01}
• training_precision: float32

Training results

Train Loss	Train Accuracy	Train Wermet	Validation Loss	Validation Accuracy	Validation Wermet	Epoch
5.2961	0.0113	1.9043	3.9402	0.0116	0.9526	0
4.6207	0.0121	0.8740	3.7957	0.0120	0.9397	1
4.4142	0.0128	0.8473	3.6045	0.0124	0.8988	2
4.1915	0.0135	0.8361	3.4445	0.0128	0.9019	3
4.0072	0.0140	0.8260	3.3268	0.0131	0.8816	4
3.8559	0.0145	0.8084	3.2440	0.0133	0.8592	5
3.7359	0.0149	0.7986	3.1751	0.0135	0.8598	6
3.6368	0.0152	0.7891	3.1298	0.0136	0.8398	7
3.5465	0.0154	0.7775	3.0736	0.0138	0.8606	8
3.4710	0.0157	0.7681	3.0318	0.0138	0.8455	9
3.3988	0.0159	0.7603	3.0159	0.0139	0.8770	10
3.3279	0.0162	0.7504	2.9672	0.0141	0.8241	11
3.2611	0.0164	0.7397	2.9541	0.0141	0.8676	12
3.1996	0.0167	0.7284	2.8913	0.0144	0.7990	13
3.1311	0.0169	0.7162	2.8671	0.0145	0.7934	14
3.0590	0.0172	0.7044	2.8241	0.0146	0.7907	15
2.9692	0.0177	0.6843	2.7517	0.0149	0.7645	16
2.8783	0.0181	0.6630	2.6682	0.0152	0.7263	17
2.7622	0.0187	0.6417	2.5586	0.0156	0.7220	18
2.6164	0.0194	0.6138	2.4121	0.0161	0.6909	19
2.4405	0.0203	0.5838	2.2417	0.0167	0.6527	20
2.2404	0.0213	0.5486	2.1401	0.0170	0.6662	21
2.0196	0.0225	0.5086	1.8907	0.0180	0.5774	22
1.7917	0.0237	0.4665	1.7073	0.0186	0.5446	23
1.5286	0.0253	0.4182	1.5139	0.0194	0.4919	24
1.2991	0.0267	0.3736	1.3605	0.0200	0.4570	25
1.1117	0.0279	0.3336	1.2304	0.0205	0.4262	26
0.9643	0.0289	0.2986	1.1387	0.0209	0.4040	27
0.8404	0.0298	0.2663	1.0514	0.0213	0.3776	28
0.7408	0.0305	0.2408	0.9883	0.0216	0.3596	29
0.6542	0.0311	0.2155	0.9281	0.0218	0.3418	30
0.5800	0.0316	0.1936	0.8801	0.0221	0.3269	31
0.5168	0.0321	0.1737	0.8401	0.0222	0.3168	32
0.4595	0.0326	0.1552	0.8071	0.0224	0.3077	33
0.4080	0.0330	0.1375	0.7825	0.0225	0.2994	34
0.3646	0.0333	0.1225	0.7550	0.0226	0.2887	35
0.3234	0.0337	0.1095	0.7369	0.0227	0.2847	36
0.2878	0.0340	0.0950	0.7270	0.0228	0.2796	37
0.2542	0.0343	0.0823	0.7096	0.0229	0.2728	38
0.2238	0.0346	0.0718	0.6963	0.0229	0.2697	39
0.1974	0.0348	0.0609	0.6857	0.0230	0.2669	40
0.1714	0.0351	0.0500	0.6843	0.0230	0.2663	41
0.1488	0.0353	0.0411	0.6770	0.0230	0.2630	42
0.1296	0.0355	0.0339	0.6754	0.0231	0.2612	43
0.1117	0.0356	0.0270	0.6702	0.0231	0.2585	44

Framework versions

• Transformers 4.33.0.dev0
• TensorFlow 2.13.0
• Tokenizers 0.13.3