whisper_syl_cv12_pad_lob100_low__0050

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: 0.0397
• Train Accuracy: 0.0362
• Train Wermet: 0.0055
• Validation Loss: 0.6611
• Validation Accuracy: 0.0232
• Validation Wermet: 0.2502
• Epoch: 49

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
5.2930	0.0113	2.0658	3.9415	0.0117	0.9401	0
4.6215	0.0121	0.8917	3.7803	0.0120	0.9294	1
4.4086	0.0128	0.8403	3.6070	0.0124	0.9223	2
4.1842	0.0135	0.8337	3.4291	0.0128	0.8867	3
3.9981	0.0141	0.8182	3.3251	0.0131	0.8750	4
3.8531	0.0145	0.8058	3.2385	0.0133	0.8699	5
3.7345	0.0149	0.7925	3.1751	0.0134	0.8665	6
3.6307	0.0152	0.7851	3.1031	0.0136	0.8507	7
3.5437	0.0155	0.7717	3.0752	0.0138	0.8286	8
3.4649	0.0157	0.7651	3.0334	0.0139	0.8417	9
3.3926	0.0159	0.7531	3.0022	0.0139	0.8413	10
3.3262	0.0162	0.7462	2.9669	0.0140	0.8264	11
3.2625	0.0164	0.7367	2.9342	0.0141	0.8520	12
3.1979	0.0166	0.7231	2.9046	0.0144	0.8196	13
3.1319	0.0169	0.7133	2.8607	0.0145	0.8026	14
3.0616	0.0172	0.7007	2.8165	0.0146	0.7788	15
2.9792	0.0176	0.6816	2.7552	0.0149	0.7643	16
2.8905	0.0180	0.6641	2.6788	0.0151	0.7473	17
2.7749	0.0186	0.6424	2.5824	0.0155	0.7241	18
2.6263	0.0193	0.6159	2.4206	0.0161	0.7047	19
2.4352	0.0203	0.5829	2.2230	0.0168	0.6500	20
2.1941	0.0216	0.5411	2.0349	0.0175	0.5980	21
1.9184	0.0231	0.4922	1.7850	0.0184	0.5659	22
1.6174	0.0249	0.4371	1.5664	0.0192	0.5081	23
1.3542	0.0265	0.3851	1.3992	0.0199	0.4690	24
1.1499	0.0278	0.3408	1.2512	0.0205	0.4299	25
0.9878	0.0288	0.3029	1.1479	0.0209	0.4013	26
0.8600	0.0297	0.2735	1.0527	0.0213	0.3755	27
0.7516	0.0305	0.2441	0.9803	0.0216	0.3570	28
0.6626	0.0311	0.2197	0.9314	0.0219	0.3416	29
0.5863	0.0316	0.1993	0.8730	0.0221	0.3238	30
0.5187	0.0321	0.1775	0.8357	0.0223	0.3136	31
0.4608	0.0326	0.1610	0.8059	0.0224	0.3033	32
0.4087	0.0330	0.1467	0.7746	0.0226	0.2949	33
0.3642	0.0334	0.1298	0.7476	0.0227	0.2847	34
0.3221	0.0337	0.1168	0.7330	0.0228	0.2802	35
0.2837	0.0340	0.1030	0.7093	0.0229	0.2728	36
0.2509	0.0343	0.0882	0.6941	0.0229	0.2687	37
0.2209	0.0346	0.0747	0.6892	0.0230	0.2656	38
0.1934	0.0349	0.0670	0.6824	0.0230	0.2630	39
0.1688	0.0351	0.0542	0.6773	0.0230	0.2625	40
0.1469	0.0353	0.0429	0.6700	0.0231	0.2633	41
0.1268	0.0355	0.0365	0.6680	0.0231	0.2578	42
0.1086	0.0357	0.0284	0.6643	0.0231	0.2540	43
0.0920	0.0358	0.0221	0.6645	0.0231	0.2530	44
0.0783	0.0359	0.0169	0.6621	0.0232	0.2540	45
0.0667	0.0360	0.0121	0.6714	0.0232	0.2532	46
0.0563	0.0361	0.0094	0.6604	0.0232	0.2503	47
0.0477	0.0361	0.0072	0.6620	0.0232	0.2489	48
0.0397	0.0362	0.0055	0.6611	0.0232	0.2502	49

Framework versions

• Transformers 4.33.0.dev0
• TensorFlow 2.13.0
• Tokenizers 0.13.3