配音擅长领域: 朗诵,情感,影视,温柔细腻,自然动听,读书
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。
训练日志
2025-07-17 16:46:48,288 男声读书 INFO {'train': {'log_interval': 100, 'eval_interval': 500, 'seed': 1234, 'epochs': 20, 'learning_rate': 0.0001, 'betas': [0.8, 0.99], 'eps': 1e-09, 'batch_size': 8, '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, 'pretrained_s2G': 'GPT_SoVITS/pretrained_models/gsv-v2final-pretrained/s2G2333k.pth', 'pretrained_s2D': 'GPT_SoVITS/pretrained_models/gsv-v2final-pretrained/s2D2333k.pth', 'if_save_latest': True, 'if_save_every_weights': True, 'save_every_epoch': 20, 'gpu_numbers': '0'}, '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/男声读书'}, 'model': {'inter_channels': 192, 'hidden_channels': 192, 'filter_channels': 768, 'n_heads': 2, 'n_layers': 6, 'kernel_size': 3, 'p_dropout': 0.1, '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': 512, 'upsample_kernel_sizes': [16, 16, 8, 2, 2], 'n_layers_q': 3, 'use_spectral_norm': False, 'gin_channels': 512, 'semantic_frame_rate': '25hz', 'freeze_quantizer': True, 'version': 'v2'}, 's2_ckpt_dir': 'logs/男声读书', 'content_module': 'cnhubert', 'save_weight_dir': 'SoVITS_weights_v2', 'name': '男声读书', 'version': 'v2', 'pretrain': None, 'resume_step': None}
2025-07-17 16:46:49,710 男声读书 INFO loaded pretrained GPT_SoVITS/pretrained_models/gsv-v2final-pretrained/s2G2333k.pth
2025-07-17 16:46:49,918 男声读书 INFO loaded pretrained GPT_SoVITS/pretrained_models/gsv-v2final-pretrained/s2D2333k.pth
2025-07-17 16:47:23,034 男声读书 INFO Train Epoch: 1 [0%]
2025-07-17 16:47:23,034 男声读书 INFO [2.8076000213623047, 1.6687934398651123, 6.665792942047119, 19.135211944580078, 0.0, 1.742974877357483, 0, 9.99875e-05]
2025-07-17 16:47:49,196 男声读书 INFO ====> Epoch: 1
2025-07-17 16:48:11,218 男声读书 INFO ====> Epoch: 2
2025-07-17 16:48:32,430 男声读书 INFO ====> Epoch: 3
2025-07-17 16:48:38,009 男声读书 INFO Train Epoch: 4 [12%]
2025-07-17 16:48:38,009 男声读书 INFO [2.6795833110809326, 2.302513599395752, 8.22059154510498, 19.794076919555664, 0.0, 2.145474672317505, 100, 9.995000937421877e-05]
2025-07-17 16:48:54,881 男声读书 INFO ====> Epoch: 4
2025-07-17 16:49:15,630 男声读书 INFO ====> Epoch: 5
2025-07-17 16:49:36,399 男声读书 INFO ====> Epoch: 6
2025-07-17 16:49:43,771 男声读书 INFO Train Epoch: 7 [25%]
2025-07-17 16:49:43,772 男声读书 INFO [2.581143379211426, 2.198768138885498, 9.208781242370605, 18.972705841064453, 0.0, 1.5633752346038818, 200, 9.991253280566489e-05]
2025-07-17 16:49:57,295 男声读书 INFO ====> Epoch: 7
2025-07-17 16:50:17,550 男声读书 INFO ====> Epoch: 8
2025-07-17 16:50:38,096 男声读书 INFO ====> Epoch: 9
2025-07-17 16:50:47,774 男声读书 INFO Train Epoch: 10 [38%]
2025-07-17 16:50:47,774 男声读书 INFO [2.701443910598755, 2.1661949157714844, 8.97201919555664, 20.136266708374023, 0.0, 1.385621190071106, 300, 9.987507028906759e-05]
2025-07-17 16:50:59,125 男声读书 INFO ====> Epoch: 10
2025-07-17 16:51:19,407 男声读书 INFO ====> Epoch: 11
2025-07-17 16:51:40,622 男声读书 INFO ====> Epoch: 12
2025-07-17 16:51:52,451 男声读书 INFO Train Epoch: 13 [50%]
2025-07-17 16:51:52,452 男声读书 INFO [2.706397533416748, 1.953716516494751, 6.676906585693359, 16.774282455444336, 0.0, 1.518265724182129, 400, 9.983762181915804e-05]
2025-07-17 16:52:01,327 男声读书 INFO ====> Epoch: 13
2025-07-17 16:52:21,659 男声读书 INFO ====> Epoch: 14
2025-07-17 16:52:42,011 男声读书 INFO ====> Epoch: 15
2025-07-17 16:53:39,293 男声读书 INFO Train Epoch: 16 [62%]
2025-07-17 16:53:39,293 男声读书 INFO [2.742380380630493, 2.278963327407837, 9.301309585571289, 19.363893508911133, 0.0, 1.0174858570098877, 500, 9.980018739066937e-05]
2025-07-17 16:53:46,055 男声读书 INFO ====> Epoch: 16
2025-07-17 16:54:06,222 男声读书 INFO ====> Epoch: 17
2025-07-17 16:54:26,902 男声读书 INFO ====> Epoch: 18
2025-07-17 16:54:42,944 男声读书 INFO Train Epoch: 19 [75%]
2025-07-17 16:54:42,944 男声读书 INFO [2.7456867694854736, 2.3350634574890137, 9.37385082244873, 18.999614715576172, 0.0, 1.0780953168869019, 600, 9.976276699833672e-05]
2025-07-17 16:54:47,322 男声读书 INFO ====> Epoch: 19
2025-07-17 16:55:07,654 男声读书 INFO Saving model and optimizer state at iteration 20 to logs/男声读书/logs_s2\G_233333333333.pth
2025-07-17 16:55:08,483 男声读书 INFO Saving model and optimizer state at iteration 20 to logs/男声读书/logs_s2\D_233333333333.pth
2025-07-17 16:55:10,454 男声读书 INFO saving ckpt 男声读书_e20:Success.
2025-07-17 16:55:10,454 男声读书 INFO ====> Epoch: 20
模型直链下载
如何使用配音模型
1,GPT-SoVITS模型GUI客户端配音
2,GPT-SoVITS模型云端部署
客服微信 xiaoming1870
暂无评论内容