人物介绍
嘉然（Diana）是乐华娱乐旗下虚拟偶像女团“A-SOUL”成员，组合吃货担当。所属组合A-SOUL是乐华娱乐2020年度企划中打造的虚拟偶像女团，共由五名成员（向晚Ava、贝拉Bella、珈乐Carol、嘉然Diana、乃琳Eileen）组成，于2020年11月23日首次公开； [1]2020年12月2日发布第一首团体单曲《Quiet》 [3]。2021年4月30日发布第二首团体单曲《超级敏感》； [4]12月17日，发布A-SOUL出道一周年系列主题曲《传说的世界》 [10]。2022年1月，获得哔哩哔哩“2021年度百大UP主”、“2021年度虚拟UP主”称号 [6] [16-17]；2月3日，发布虎年贺岁单曲《除夕》 [9]。
截止2022年12月9日，嘉然在哔哩哔哩收获了169.3万粉丝
配音模型效果
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
2024-07-02 10:32:44,574 jiaran INFO {'train': {'log_interval': 100, 'eval_interval': 500, 'seed': 1234, 'epochs': 8, '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/s2G488k.pth', 'pretrained_s2D': 'GPT_SoVITS/pretrained_models/s2D488k.pth', 'if_save_latest': True, 'if_save_every_weights': True, 'save_every_epoch': 4, '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/jiaran'}, '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}, 's2_ckpt_dir': 'logs/jiaran', 'content_module': 'cnhubert', 'save_weight_dir': 'SoVITS_weights', 'name': 'jiaran', 'pretrain': None, 'resume_step': None}
2024-07-02 10:32:45,960 jiaran INFO loaded pretrained GPT_SoVITS/pretrained_models/s2G488k.pth
2024-07-02 10:32:46,156 jiaran INFO loaded pretrained GPT_SoVITS/pretrained_models/s2D488k.pth
2024-07-02 10:33:15,022 jiaran INFO Train Epoch: 1 [0%]
2024-07-02 10:33:15,022 jiaran INFO [2.4960834980010986, 2.2735910415649414, 7.0620341300964355, 19.9090576171875, 0.0, 2.9329488277435303, 0, 9.99875e-05]
2024-07-02 10:33:43,061 jiaran 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': 20, '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/s2G488k.pth', 'pretrained_s2D': 'GPT_SoVITS/pretrained_models/s2D488k.pth', 'if_save_latest': True, 'if_save_every_weights': True, 'save_every_epoch': 10, '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/jiaran'}, '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}, 's2_ckpt_dir': 'logs/jiaran', 'content_module': 'cnhubert', 'save_weight_dir': 'SoVITS_weights', 'name': 'jiaran', 'pretrain': None, 'resume_step': None}
2024-07-02 10:33:44,296 jiaran INFO loaded pretrained GPT_SoVITS/pretrained_models/s2G488k.pth
2024-07-02 10:33:44,510 jiaran INFO loaded pretrained GPT_SoVITS/pretrained_models/s2D488k.pth
2024-07-02 23:15:09,755 jiaran INFO Train Epoch: 1 [0%]
2024-07-02 23:15:09,755 jiaran INFO [2.603549003601074, 2.170898914337158, 9.015737533569336, 20.780746459960938, 0.0, 2.8269424438476562, 0, 9.99875e-05]
2024-07-02 23:15:24,270 jiaran INFO ====> Epoch: 1
2024-07-02 23:15:38,422 jiaran INFO ====> Epoch: 2
2024-07-02 23:15:52,663 jiaran INFO ====> Epoch: 3
2024-07-02 23:19:07,019 jiaran INFO ====> Epoch: 4
2024-07-02 23:19:20,886 jiaran INFO ====> Epoch: 5
2024-07-02 23:19:34,134 jiaran INFO Train Epoch: 6 [88%]
2024-07-02 23:19:34,134 jiaran INFO [2.6198172569274902, 2.351412773132324, 9.823515892028809, 19.64070701599121, 0.0, 1.6281896829605103, 100, 9.99250234335941e-05]
2024-07-02 23:19:35,246 jiaran INFO ====> Epoch: 6
2024-07-02 23:19:49,363 jiaran INFO ====> Epoch: 7
2024-07-02 10:36:03,333 jiaran INFO ====> Epoch: 8
2024-07-02 10:36:17,042 jiaran INFO ====> Epoch: 9
2024-07-02 10:36:31,009 jiaran INFO Saving model and optimizer state at iteration 10 to logs/jiaran/logs_s2\G_233333333333.pth
2024-07-02 10:36:31,829 jiaran INFO Saving model and optimizer state at iteration 10 to logs/jiaran/logs_s2\D_233333333333.pth
2024-07-02 10:36:32,947 jiaran INFO saving ckpt jiaran_e10:Success.
2024-07-02 10:36:32,948 jiaran INFO ====> Epoch: 10
2024-07-02 10:36:47,323 jiaran INFO ====> Epoch: 11
2024-07-02 10:36:59,202 jiaran INFO Train Epoch: 12 [76%]
2024-07-02 10:36:59,202 jiaran INFO [2.5783495903015137, 2.3862318992614746, 10.590505599975586, 19.603647232055664, 0.0, 1.4245914220809937, 200, 9.98501030820433e-05]
2024-07-02 10:37:01,817 jiaran INFO ====> Epoch: 12
2024-07-02 10:37:15,825 jiaran INFO ====> Epoch: 13
2024-07-02 10:37:29,798 jiaran INFO ====> Epoch: 14
2024-07-02 10:37:43,754 jiaran INFO ====> Epoch: 15
2024-07-02 10:37:57,609 jiaran INFO ====> Epoch: 16
2024-07-02 10:38:11,718 jiaran INFO ====> Epoch: 17
2024-07-02 10:38:22,118 jiaran INFO Train Epoch: 18 [65%]
2024-07-02 10:38:22,119 jiaran INFO [2.4857614040374756, 2.3586795330047607, 9.014538764953613, 19.638473510742188, 0.0, 1.8002816438674927, 300, 9.977523890319963e-05]
2024-07-02 10:38:26,170 jiaran INFO ====> Epoch: 18
2024-07-02 10:38:40,110 jiaran INFO ====> Epoch: 19
2024-07-02 10:38:54,206 jiaran INFO Saving model and optimizer state at iteration 20 to logs/jiaran/logs_s2\G_233333333333.pth
2024-07-02 10:38:55,269 jiaran INFO Saving model and optimizer state at iteration 20 to logs/jiaran/logs_s2\D_233333333333.pth
2024-07-02 10:38:57,732 jiaran INFO saving ckpt jiaran_e20:Success.
2024-07-02 10:38:57,732 jiaran 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