GPT-SOVITS模型|温柔男声声音模型

擅长领域   磁性浑厚,自然动听 |情感,小说,影视,动漫,纪录片,百科,历史

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

训练日志

2024-08-15 16:58:32,633	peiyin.me 温柔男声1	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 温柔男声1'}, '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 温柔男声1', 'content_module': 'cnhubert', 'save_weight_dir': 'SoVITS_weights_v2', 'name': 'peiyin.me 温柔男声1', 'version': 'v2', 'pretrain': None, 'resume_step': None}
2024-08-15 16:58:34,904	peiyin.me 温柔男声1	INFO	loaded pretrained GPT_SoVITS/pretrained_models/gsv-v2final-pretrained/s2G2333k.pth
2024-08-15 16:58:35,125	peiyin.me 温柔男声1	INFO	loaded pretrained GPT_SoVITS/pretrained_models/gsv-v2final-pretrained/s2D2333k.pth
2024-08-15 16:59:25,574	peiyin.me 温柔男声1	INFO	Train Epoch: 1 [0%]
2024-08-15 16:59:25,575	peiyin.me 温柔男声1	INFO	[2.6815168857574463, 1.557905673980713, 7.528033256530762, 23.13617515563965, 0.0, 2.3211686611175537, 0, 9.99875e-05]
2024-08-15 16:59:42,958	peiyin.me 温柔男声1	INFO	====> Epoch: 1
2024-08-15 16:59:57,647	peiyin.me 温柔男声1	INFO	====> Epoch: 2
2024-08-15 17:00:11,488	peiyin.me 温柔男声1	INFO	====> Epoch: 3
2024-08-15 17:00:25,301	peiyin.me 温柔男声1	INFO	====> Epoch: 4
2024-08-15 17:00:39,242	peiyin.me 温柔男声1	INFO	====> Epoch: 5
2024-08-15 17:00:41,249	peiyin.me 温柔男声1	INFO	Train Epoch: 6 [0%]
2024-08-15 17:00:41,250	peiyin.me 温柔男声1	INFO	[2.511470317840576, 2.283311605453491, 7.676183700561523, 20.58940887451172, 0.0, 1.5977298021316528, 100, 9.99250234335941e-05]
2024-08-15 17:00:53,959	peiyin.me 温柔男声1	INFO	====> Epoch: 6
2024-08-15 17:01:08,133	peiyin.me 温柔男声1	INFO	====> Epoch: 7
2024-08-15 17:01:22,241	peiyin.me 温柔男声1	INFO	====> Epoch: 8
2024-08-15 17:01:41,193	peiyin.me 温柔男声1	INFO	====> Epoch: 9
2024-08-15 17:02:04,447	peiyin.me 温柔男声1	INFO	====> Epoch: 10
2024-08-15 17:02:07,212	peiyin.me 温柔男声1	INFO	Train Epoch: 11 [0%]
2024-08-15 17:02:07,212	peiyin.me 温柔男声1	INFO	[2.444606304168701, 2.353349208831787, 8.39655876159668, 22.016489028930664, 0.0, 1.4380923509597778, 200, 9.986258590528146e-05]
2024-08-15 17:02:27,878	peiyin.me 温柔男声1	INFO	====> Epoch: 11
2024-08-15 17:02:41,681	peiyin.me 温柔男声1	INFO	====> Epoch: 12
2024-08-15 17:02:55,860	peiyin.me 温柔男声1	INFO	====> Epoch: 13
2024-08-15 17:03:09,408	peiyin.me 温柔男声1	INFO	====> Epoch: 14
2024-08-15 17:03:23,055	peiyin.me 温柔男声1	INFO	====> Epoch: 15
2024-08-15 17:03:25,126	peiyin.me 温柔男声1	INFO	Train Epoch: 16 [0%]
2024-08-15 17:03:25,127	peiyin.me 温柔男声1	INFO	[2.4419913291931152, 2.573406934738159, 8.30177116394043, 20.688777923583984, 0.0, 1.4545917510986328, 300, 9.980018739066937e-05]
2024-08-15 17:03:37,564	peiyin.me 温柔男声1	INFO	====> Epoch: 16
2024-08-15 17:03:51,494	peiyin.me 温柔男声1	INFO	====> Epoch: 17
2024-08-15 17:04:05,520	peiyin.me 温柔男声1	INFO	====> Epoch: 18
2024-08-15 17:04:19,235	peiyin.me 温柔男声1	INFO	====> Epoch: 19
2024-08-15 17:04:33,185	peiyin.me 温柔男声1	INFO	Saving model and optimizer state at iteration 20 to logs/peiyin.me 温柔男声1/logs_s2\G_233333333333.pth
2024-08-15 17:04:34,164	peiyin.me 温柔男声1	INFO	Saving model and optimizer state at iteration 20 to logs/peiyin.me 温柔男声1/logs_s2\D_233333333333.pth
2024-08-15 17:04:36,363	peiyin.me 温柔男声1	INFO	saving ckpt peiyin.me 温柔男声1_e20:Success.
2024-08-15 17:04:36,363	peiyin.me 温柔男声1	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。

请登录后发表评论

    没有回复内容