helldivers2-jarvis-asrV8

This model is a fine-tuned version of facebook/wav2vec2-base-960h on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 19.1922
• Wer: 0.1770
• Cer: 0.7950

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: 1e-05
• train_batch_size: 15
• eval_batch_size: 15
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 50
• num_epochs: 70

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
585.4988	1.0	73	271.5495	0.3633	0.8109
378.3453	2.0	146	168.1931	0.3276	0.8056
319.9564	3.0	219	140.2753	0.3131	0.8034
281.8407	4.0	292	127.9118	0.2933	0.8023
264.2254	5.0	365	103.9656	0.2814	0.8020
233.3232	6.0	438	87.8603	0.2655	0.8005
213.0622	7.0	511	80.5251	0.2497	0.7993
194.7588	8.0	584	77.2796	0.2457	0.7991
197.9257	9.0	657	72.9397	0.2378	0.7985
183.4031	10.0	730	57.6201	0.2312	0.7979
172.3133	11.0	803	56.8209	0.2219	0.7976
165.2571	12.0	876	52.5103	0.2219	0.7972
161.9286	13.0	949	50.8869	0.2180	0.7970
158.4977	14.0	1022	46.2574	0.2127	0.7969
146.6963	15.0	1095	39.4797	0.2114	0.7968
131.5992	16.0	1168	38.5002	0.2127	0.7966
136.5205	17.0	1241	34.6582	0.2100	0.7965
130.8528	18.0	1314	31.8192	0.2061	0.7964
122.9594	19.0	1387	31.4980	0.1968	0.7961
122.6427	20.0	1460	29.6717	0.1955	0.7961
116.4709	21.0	1533	30.8555	0.1955	0.7960
124.5335	22.0	1606	31.9685	0.1942	0.7959
110.134	23.0	1679	32.0465	0.1968	0.7963
105.5423	24.0	1752	30.3238	0.1955	0.7961
113.3944	25.0	1825	28.7380	0.1889	0.7959
112.8572	26.0	1898	25.8070	0.1876	0.7957
106.1037	27.0	1971	28.3786	0.1876	0.7957
105.8874	28.0	2044	27.5647	0.1889	0.7957
96.3391	29.0	2117	29.5317	0.1915	0.7958
103.2364	30.0	2190	26.4454	0.1836	0.7956
101.3744	31.0	2263	24.9761	0.1836	0.7955
94.1529	32.0	2336	25.9349	0.1810	0.7954
89.4031	33.0	2409	18.5316	0.1810	0.7954
98.6291	34.0	2482	26.2887	0.1810	0.7954
88.7591	35.0	2555	26.2593	0.1823	0.7954
91.8604	36.0	2628	27.2263	0.1823	0.7954
83.4981	37.0	2701	26.0174	0.1810	0.7954
88.0891	38.0	2774	26.2153	0.1823	0.7955
85.7155	39.0	2847	24.1102	0.1810	0.7954
87.0231	40.0	2920	25.7208	0.1836	0.7955
85.2278	41.0	2993	25.6799	0.1797	0.7952
83.8083	42.0	3066	25.9679	0.1810	0.7955
83.1864	43.0	3139	23.0670	0.1797	0.7953
92.3624	44.0	3212	23.0629	0.1797	0.7953
101.0567	45.0	3285	23.7475	0.1823	0.7955
77.1269	46.0	3358	21.2927	0.1836	0.7954
79.782	47.0	3431	22.2427	0.1797	0.7952
79.0065	48.0	3504	22.0512	0.1783	0.7952
78.0512	49.0	3577	22.9018	0.1770	0.7950
84.4353	50.0	3650	23.3538	0.1783	0.7952
76.3221	51.0	3723	21.3755	0.1823	0.7952
74.6622	52.0	3796	22.8240	0.1783	0.7951
82.2913	53.0	3869	21.2273	0.1770	0.7950
71.0879	54.0	3942	20.8637	0.1770	0.7950
78.6892	55.0	4015	22.2313	0.1770	0.7951
68.3446	56.0	4088	20.8211	0.1797	0.7952
78.6938	57.0	4161	21.4202	0.1783	0.7952
76.9933	58.0	4234	20.0210	0.1783	0.7951
70.8471	59.0	4307	22.2786	0.1783	0.7951
76.5866	60.0	4380	20.0782	0.1770	0.7951
69.9045	61.0	4453	21.1576	0.1797	0.7951
81.1508	62.0	4526	18.7841	0.1797	0.7951
82.9743	63.0	4599	18.4739	0.1823	0.7952
80.8333	64.0	4672	20.2582	0.1810	0.7954
71.9855	65.0	4745	21.3909	0.1770	0.7952
80.0273	66.0	4818	20.2303	0.1797	0.7952
74.9079	67.0	4891	19.8304	0.1783	0.7952
75.7448	68.0	4964	21.5770	0.1823	0.7954
71.0606	69.0	5037	20.9629	0.1797	0.7952
75.027	70.0	5110	19.1922	0.1770	0.7950

Framework versions

• Transformers 4.51.3
• Pytorch 2.5.1+cu121
• Datasets 3.6.0
• Tokenizers 0.21.1