声音特点适用领域：成熟知性，温婉大方，情感，名人，影视，动漫，小说
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。
模型下载
https://vip.123pan.cn/1816369016/13776108
训练日志
2024-08-10 13:37:01,886 peiyin.me_女声_理性_新闻 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': 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, '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/peiyin.me_女声_理性_新闻'}, '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/peiyin.me_女声_理性_新闻', 'content_module': 'cnhubert', 'save_weight_dir': 'SoVITS_weights_v2', 'name': 'peiyin.me_女声_理性_新闻', 'version': 'v2', 'pretrain': None, 'resume_step': None}
2024-08-10 13:37:03,490 peiyin.me_女声_理性_新闻 INFO loaded pretrained GPT_SoVITS/pretrained_models/gsv-v2final-pretrained/s2G2333k.pth
2024-08-10 13:37:03,792 peiyin.me_女声_理性_新闻 INFO loaded pretrained GPT_SoVITS/pretrained_models/gsv-v2final-pretrained/s2D2333k.pth
2024-08-10 13:37:47,751 peiyin.me_女声_理性_新闻 INFO Train Epoch: 1 [0%]
2024-08-10 13:37:47,752 peiyin.me_女声_理性_新闻 INFO [3.0324933528900146, 1.554208755493164, 7.521521091461182, 24.222415924072266, 0.0, 2.175811767578125, 0, 9.99875e-05]
2024-08-10 13:38:10,497 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 1
2024-08-10 13:38:34,249 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 2
2024-08-10 13:38:56,866 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 3
2024-08-10 13:39:19,616 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 4
2024-08-10 13:39:41,577 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 5
2024-08-10 13:40:02,501 peiyin.me_女声_理性_新闻 INFO Train Epoch: 6 [88%]
2024-08-10 13:40:02,511 peiyin.me_女声_理性_新闻 INFO [2.7591938972473145, 2.107762098312378, 9.227492332458496, 24.212217330932617, 0.0, 1.7909579277038574, 100, 9.99250234335941e-05]
2024-08-10 13:40:04,120 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 6
2024-08-10 13:40:26,281 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 7
2024-08-10 13:40:48,986 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 8
2024-08-10 13:41:11,007 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 9
2024-08-10 13:41:33,804 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 10
2024-08-10 13:41:55,722 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 11
2024-08-10 13:42:14,948 peiyin.me_女声_理性_新闻 INFO Train Epoch: 12 [76%]
2024-08-10 13:42:14,963 peiyin.me_女声_理性_新闻 INFO [2.6518325805664062, 2.2211127281188965, 9.298238754272461, 25.66077423095703, 0.0, 1.359619379043579, 200, 9.98501030820433e-05]
2024-08-10 13:42:18,698 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 12
2024-08-10 13:42:41,803 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 13
2024-08-10 13:43:04,239 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 14
2024-08-10 13:43:27,349 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 15
2024-08-10 13:43:49,821 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 16
2024-08-10 13:44:12,615 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 17
2024-08-10 13:44:29,487 peiyin.me_女声_理性_新闻 INFO Train Epoch: 18 [65%]
2024-08-10 13:44:29,493 peiyin.me_女声_理性_新闻 INFO [2.6113979816436768, 2.2373108863830566, 10.365253448486328, 24.591243743896484, 0.0, 1.9521478414535522, 300, 9.977523890319963e-05]
2024-08-10 13:44:35,247 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 18
2024-08-10 13:44:57,100 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 19
2024-08-10 13:45:18,762 peiyin.me_女声_理性_新闻 INFO Saving model and optimizer state at iteration 20 to logs/peiyin.me_女声_理性_新闻/logs_s2\G_233333333333.pth
2024-08-10 13:45:19,869 peiyin.me_女声_理性_新闻 INFO Saving model and optimizer state at iteration 20 to logs/peiyin.me_女声_理性_新闻/logs_s2\D_233333333333.pth
2024-08-10 13:45:21,648 peiyin.me_女声_理性_新闻 INFO saving ckpt peiyin.me_女声_理性_新闻_e20:Success.
2024-08-10 13:45:21,648 peiyin.me_女声_理性_新闻 INFO ====> Epoch: 20如何使用配音模型
1，gpt-sovits模型云端部署
https://aiaf.cc/gpt-sovits-yunduan/.html
2，gpt-sovits模型本地部署
https://aiaf.cc/gpt-sovits/.html
如果您想一对一远程教学模型安装、模型训练，请联系微信 xiaoming1870
声音版权使用声明
本网站展示的 AI 声音模型由站长及工作室精心创作并提供。遵循非商业性使用原则，仅作娱乐用途，重视并遵守版权所有者权益，未获授权也不声称拥有使用权。模型整理等产生的费用仅覆盖服务成本，不涉及版权收费。所有活动在法律框架内进行，尊重版权、合法使用分享。如有疑问、需版权信息或建议反馈，可随时联系，共同促进 AI 声音艺术发展与营造尊重版权氛围。
没有回复内容