GPT-SoVITS模型|成熟中年纪录片、成熟中年男声配音

浑厚大气，讲述自然浑厚的中年男声，物超所值! 永久授权，本地离线模型

GPT-SoVITS模型配音效果

鉴于GPT-SoVITS模型自回归特性，即其配音情绪高度依赖于所提供的参考音频，特此说明：本视频所展示的配音情绪仅为采用某一特定参考音频时的效果示例，并不全面反映GPT-SoVITS模型能够生成的全部情绪范围及最终配音质量的上限。模型的最终表现将随着不同参考音频的输入而展现出多样化。

融入大自然的怀抱，感受生命的韵律。

纪录片中，我们捕捉到了世界上最珍贵的瞬间。

模型下载

训练日志

2025-09-24 20:58:12,140	成熟中年纪录片	INFO	{'train': {'log_interval': 100, 'eval_interval': 500, 'seed': 1234, 'epochs': 15, 'learning_rate': 0.0001, 'betas': [0.8, 0.99], 'eps': 1e-09, 'batch_size': 11, 'fp16_run': True, 'lr_decay': 0.999875, 'segment_size': 20480, 'init_lr_ratio': 1, 'warmup_epochs': 0, 'c_mel': 45, 'c_kl': 1.0, 'text_low_lr_rate': 0.4, 'grad_ckpt': False, 'pretrained_s2G': 'GPT_SoVITS/pretrained_models/v2Pro/s2Gv2ProPlus.pth', 'pretrained_s2D': 'GPT_SoVITS/pretrained_models/v2Pro/s2Dv2ProPlus.pth', 'if_save_latest': True, 'if_save_every_weights': True, 'save_every_epoch': 15, 'gpu_numbers': '0', 'lora_rank': '32'}, 'data': {'max_wav_value': 32768.0, 'sampling_rate': 32000, 'filter_length': 2048, 'hop_length': 640, 'win_length': 2048, 'n_mel_channels': 128, 'mel_fmin': 0.0, 'mel_fmax': None, 'add_blank': True, 'n_speakers': 300, 'cleaned_text': True, 'exp_dir': 'logs/成熟中年纪录片'}, 'model': {'inter_channels': 192, 'hidden_channels': 192, 'filter_channels': 768, 'n_heads': 2, 'n_layers': 6, 'kernel_size': 3, 'p_dropout': 0.0, 'resblock': '1', 'resblock_kernel_sizes': [3, 7, 11], 'resblock_dilation_sizes': [[1, 3, 5], [1, 3, 5], [1, 3, 5]], 'upsample_rates': [10, 8, 2, 2, 2], 'upsample_initial_channel': 768, 'upsample_kernel_sizes': [20, 16, 8, 2, 2], 'n_layers_q': 3, 'use_spectral_norm': False, 'gin_channels': 1024, 'semantic_frame_rate': '25hz', 'freeze_quantizer': True, 'version': 'v2ProPlus'}, 's2_ckpt_dir': 'logs/成熟中年纪录片', 'content_module': 'cnhubert', 'save_weight_dir': 'SoVITS_weights_v2ProPlus', 'name': '成熟中年纪录片', 'version': 'v2ProPlus', 'pretrain': None, 'resume_step': None}
2025-09-24 21:00:37,990	成熟中年纪录片	INFO	{'train': {'log_interval': 100, 'eval_interval': 500, 'seed': 1234, 'epochs': 15, 'learning_rate': 0.0001, 'betas': [0.8, 0.99], 'eps': 1e-09, 'batch_size': 11, 'fp16_run': True, 'lr_decay': 0.999875, 'segment_size': 20480, 'init_lr_ratio': 1, 'warmup_epochs': 0, 'c_mel': 45, 'c_kl': 1.0, 'text_low_lr_rate': 0.4, 'grad_ckpt': False, 'pretrained_s2G': 'GPT_SoVITS/pretrained_models/v2Pro/s2Gv2ProPlus.pth', 'pretrained_s2D': 'GPT_SoVITS/pretrained_models/v2Pro/s2Dv2ProPlus.pth', 'if_save_latest': True, 'if_save_every_weights': True, 'save_every_epoch': 15, 'gpu_numbers': '0', 'lora_rank': '32'}, 'data': {'max_wav_value': 32768.0, 'sampling_rate': 32000, 'filter_length': 2048, 'hop_length': 640, 'win_length': 2048, 'n_mel_channels': 128, 'mel_fmin': 0.0, 'mel_fmax': None, 'add_blank': True, 'n_speakers': 300, 'cleaned_text': True, 'exp_dir': 'logs/成熟中年纪录片'}, 'model': {'inter_channels': 192, 'hidden_channels': 192, 'filter_channels': 768, 'n_heads': 2, 'n_layers': 6, 'kernel_size': 3, 'p_dropout': 0.0, 'resblock': '1', 'resblock_kernel_sizes': [3, 7, 11], 'resblock_dilation_sizes': [[1, 3, 5], [1, 3, 5], [1, 3, 5]], 'upsample_rates': [10, 8, 2, 2, 2], 'upsample_initial_channel': 768, 'upsample_kernel_sizes': [20, 16, 8, 2, 2], 'n_layers_q': 3, 'use_spectral_norm': False, 'gin_channels': 1024, 'semantic_frame_rate': '25hz', 'freeze_quantizer': True, 'version': 'v2ProPlus'}, 's2_ckpt_dir': 'logs/成熟中年纪录片', 'content_module': 'cnhubert', 'save_weight_dir': 'SoVITS_weights_v2ProPlus', 'name': '成熟中年纪录片', 'version': 'v2ProPlus', 'pretrain': None, 'resume_step': None}
2025-09-24 21:00:40,120	成熟中年纪录片	INFO	loaded pretrained GPT_SoVITS/pretrained_models/v2Pro/s2Gv2ProPlus.pth
2025-09-24 21:00:40,599	成熟中年纪录片	INFO	loaded pretrained GPT_SoVITS/pretrained_models/v2Pro/s2Dv2ProPlus.pth
2025-09-24 21:01:26,374	成熟中年纪录片	INFO	Train Epoch: 1 [0%]
2025-09-24 21:01:26,374	成熟中年纪录片	INFO	[4.279621124267578, 1.655268907546997, 17.71141242980957, 26.8634090423584, 0.0, 3.357395648956299, 0, 9.99875e-05]
2025-09-24 21:01:51,263	成熟中年纪录片	INFO	====> Epoch: 1
2025-09-24 21:02:15,616	成熟中年纪录片	INFO	====> Epoch: 2
2025-09-24 21:02:39,433	成熟中年纪录片	INFO	====> Epoch: 3
2025-09-24 21:03:01,036	成熟中年纪录片	INFO	Train Epoch: 4 [85%]
2025-09-24 21:03:01,037	成熟中年纪录片	INFO	[3.7005553245544434, 2.6369876861572266, 14.739365577697754, 21.873214721679688, 0.0, 1.445163607597351, 100, 9.995000937421877e-05]
2025-09-24 21:03:03,712	成熟中年纪录片	INFO	====> Epoch: 4
2025-09-24 21:03:27,440	成熟中年纪录片	INFO	====> Epoch: 5
2025-09-24 21:03:51,150	成熟中年纪录片	INFO	====> Epoch: 6
2025-09-24 21:04:14,969	成熟中年纪录片	INFO	====> Epoch: 7
2025-09-24 21:04:33,872	成熟中年纪录片	INFO	Train Epoch: 8 [69%]
2025-09-24 21:04:33,872	成熟中年纪录片	INFO	[3.8219761848449707, 2.7108981609344482, 12.585599899291992, 20.52610206604004, 0.0, 0.989400327205658, 200, 9.990004373906418e-05]
2025-09-24 21:04:39,647	成熟中年纪录片	INFO	====> Epoch: 8
2025-09-24 21:05:04,540	成熟中年纪录片	INFO	====> Epoch: 9
2025-09-24 21:05:34,005	成熟中年纪录片	INFO	====> Epoch: 10
2025-09-24 21:05:59,371	成熟中年纪录片	INFO	====> Epoch: 11
2025-09-24 21:06:15,542	成熟中年纪录片	INFO	Train Epoch: 12 [54%]
2025-09-24 21:06:15,543	成熟中年纪录片	INFO	[3.705124616622925, 2.7097651958465576, 18.957538604736328, 23.957239151000977, 0.0, 1.157300591468811, 300, 9.98501030820433e-05]
2025-09-24 21:06:24,765	成熟中年纪录片	INFO	====> Epoch: 12
2025-09-24 21:06:51,884	成熟中年纪录片	INFO	====> Epoch: 13
2025-09-24 21:07:17,229	成熟中年纪录片	INFO	====> Epoch: 14
2025-09-24 21:07:44,242	成熟中年纪录片	INFO	Saving model and optimizer state at iteration 15 to logs/成熟中年纪录片/logs_s2_v2ProPlus\G_233333333333.pth
2025-09-24 21:07:49,163	成熟中年纪录片	INFO	Saving model and optimizer state at iteration 15 to logs/成熟中年纪录片/logs_s2_v2ProPlus\D_233333333333.pth
2025-09-24 21:08:07,802	成熟中年纪录片	INFO	saving ckpt 成熟中年纪录片_e15:Success.
2025-09-24 21:08:07,802	成熟中年纪录片	INFO	====> Epoch: 15