babysor/MockingBird issues and pull requests

#697 - encoder 训练 loss不降问题

Issue - State: open - Opened by dennishuang85 over 2 years ago - 2 comments

#684 - 调整完batch_size之后需要其他的操作吗？

Issue - State: closed - Opened by yunqi777 over 2 years ago - 1 comment

#677 - 使用中文readme中的第三个模型，同样的input音频文件和文字内容，得到的结果音频有杂音且每次效果不同

Issue - State: open - Opened by Jinkeee over 2 years ago - 4 comments

#665 - hifigan的权重放到官方hifigan代码里需要做什么修改吗

Issue - State: closed - Opened by xxy-im over 2 years ago

#660 - Env update 添加环境需求注释

Pull Request - State: closed - Opened by earmer over 2 years ago

#658 - 训练后为什么会变成这样？

Issue - State: open - Opened by showtime12345 over 2 years ago - 4 comments

#613 - Web server 无法跑起来

Issue - State: closed - Opened by jearton over 2 years ago - 7 comments

#607 - vc模式下电音（模型兼容性）以及24khifi_仅仅只有16000hz的问题，以及对要变声的输入hz的自适应和输出是否能达到41000hz的假设

Issue - State: open - Opened by EnderOcelot over 2 years ago - 4 comments
Labels: documentation, enhancement

#601 - No matching distribution found for numpy==1.19.3

Issue - State: open - Opened by KasuganoSora-desu over 2 years ago - 2 comments

#588 - 大佬们能来个新的进群路径吗，微信，qq都可以，搞来搞去还是杂音

Issue - State: open - Opened by skyflym over 2 years ago - 5 comments

#581 - 使用版本7f799d322f7599跑模型pretrained-11-7-21_75k.pt时遇到报错：raise ValueError("Axis limits cannot be NaN or Inf")

Issue - State: closed - Opened by kate123wong over 2 years ago - 3 comments

#555 - closed

Issue - State: closed - Opened by zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz-z over 2 years ago

#537 - 有无大佬看看这是啥问题

Issue - State: closed - Opened by SchweitzerGAO over 2 years ago - 5 comments

#521 - 使用hifigan_24k报错 KeyError: 'model_state'

Issue - State: open - Opened by chn-lee-yumi over 2 years ago - 6 comments

#504 - 训练的时候出问题 RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format( RuntimeError: Error(s) in loading state_dict for Tacotron:

Issue - State: open - Opened by glegender over 2 years ago - 2 comments

#496 - librosa reports warning as "From version 0.10 passing these as positional arguments will result in an error" that massed up the console output.

Issue - State: closed - Opened by moosewoler over 2 years ago - 5 comments

#477 - 替换模型训练遇到的问题

Issue - State: open - Opened by 1135126802 over 2 years ago - 9 comments

#462 - 运行报错缺少pyworld，安装visualstudio后pip install pyworld后报错如下，求解。

Issue - State: closed - Opened by JOKERSLION over 2 years ago - 4 comments

#437 - 【长期】训练克隆特定人声音&finetune

Issue - State: open - Opened by babysor over 2 years ago - 18 comments
Labels: discussion

#432 - 使用新版本遇到问题

Issue - State: closed - Opened by Chopin68 over 2 years ago - 14 comments

#431 - 能否部署到树莓派运行

Issue - State: open - Opened by YinJiake over 2 years ago - 4 comments

#422 - 求助！！！在下载剩余的包 pip install -r requirements.txt 报错了大佬知道咋解决吗

Issue - State: closed - Opened by frankl07 over 2 years ago - 11 comments

#395 - 前所未有的问题……预处理时显示到100%，但是什么都没生成，什么情况啊

Issue - State: closed - Opened by yy35959199 almost 3 years ago - 5 comments

#390 - 数据集预处理时报错

Issue - State: closed - Opened by yrsn509 almost 3 years ago - 3 comments

#358 - 安卓📱Android成功运行本项目

Issue - State: open - Opened by ferretgeek almost 3 years ago - 2 comments

#346 - 是否可以支持DirectML pytorch-directml?

Issue - State: open - Opened by YahuiWong almost 3 years ago - 1 comment

#311 - 使用预置模型，Toobox输出的是电音

Issue - State: closed - Opened by dewjiguang almost 3 years ago - 4 comments

#296 - modified tacotron synthesizer to tacotron2

Pull Request - State: open - Opened by Rita-ritally almost 3 years ago - 2 comments

#289 - 分享两个训练好的synthesizer模型

Issue - State: open - Opened by wrk226 almost 3 years ago - 13 comments
Labels: documentation, enhancement

#285 - RuntimeError: Error(s) in loading state_dict for Tacotron: size mismatch for gst.stl.attention.W_query.weight: copying a param with shape torch.Size([512, 256]) from checkpoint, the shape in current model is torch.Size([512, 512]).

Issue - State: closed - Opened by DanMerry almost 3 years ago - 5 comments

#271 - 进行音频和梅尔频谱图预处理报错怎么回事： python encoder_preprocess.py <datasets_root>

Issue - State: open - Opened by LiangChenStart almost 3 years ago - 7 comments

#267 - 如何平稳的停止训练，手动中断训练后再次训练报错RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory

Issue - State: closed - Opened by luobingit almost 3 years ago - 2 comments

#267 - 如何平稳的停止训练，手动中断训练后再次训练报错RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory

Issue - State: closed - Opened by luobingit almost 3 years ago - 2 comments

#257 - 怎么在Google Colab上运行？（在不想使用Toolbox和Web的情况下）

Issue - State: open - Opened by joeynmq almost 3 years ago - 4 comments
Labels: documentation

#252 - 可不可以添加一个inference.py，在指定一系列dir or文件名后就生成最终的音频呢？

Issue - State: open - Opened by zeng-yifei almost 3 years ago - 4 comments

#245 - 第二次贡献模型，同时谈谈相关经验。

Issue - State: open - Opened by ferretgeek almost 3 years ago - 13 comments

#239 - 分享600kstep模型

Issue - State: open - Opened by Yuqi154 almost 3 years ago - 21 comments

#216 - 载入模型pretrained-11-7-21_75k.pt。报错：size mismatch for gst.stl.attention.W_query.weight: copying a param with shape torch.Size([512, 512]) from checkpoint, the shape in current model is torch.Size([512, 256]).

Issue - State: closed - Opened by pyli0628 about 3 years ago - 7 comments

#131 - RuntimeError: PytorchStreamReader failed reading file data/39: invalid header or archive is corrupted

Issue - State: open - Opened by dazhao233 about 3 years ago - 5 comments

#116 - Exception: No synthesizer models found in synthesizer\saved_models

Issue - State: closed - Opened by Jason-hu-jav about 3 years ago - 5 comments

#80 - 合成失败

Issue - State: closed - Opened by ajunbudou about 3 years ago - 5 comments

#75 - Temp to support Mozilla Common Voice dataset

Pull Request - State: closed - Opened by FawenYo about 3 years ago - 2 comments

#56 - 200zh数据集解压后，第一步预处理报错

Issue - State: closed - Opened by pythonerrnewer over 3 years ago - 10 comments

#37 - 用这里的模型跑出现这个RuntimeError: Error(s) in loading state_dict for Tacotron: size mismatch for encoder.embedding.weight: copying a param with shape torch.Size([70, 512]) from checkpoint, the shape in current model is torch.Size([75, 512]).

Issue - State: closed - Opened by wangkewk over 3 years ago - 79 comments
Labels: bug, wontfix

#37 - 用这里的模型跑出现这个RuntimeError: Error(s) in loading state_dict for Tacotron: size mismatch for encoder.embedding.weight: copying a param with shape torch.Size([70, 512]) from checkpoint, the shape in current model is torch.Size([75, 512]).

Issue - State: closed - Opened by wangkewk over 3 years ago - 78 comments
Labels: bug, wontfix

#20 - 保姆级别教程（持续更新各类社区/非官方教程----

Issue - State: open - Opened by zhuzaileiting over 3 years ago - 145 comments

#10 - 想要支持更多数据集？在这里提建议

Issue - State: open - Opened by babysor over 3 years ago - 40 comments

GitHub / babysor/MockingBird issues and pull requests