issues
search
FunAudioLLM
/
CosyVoice
Multi-lingual large voice generation model, providing inference, training and deployment full-stack ability.
https://funaudiollm.github.io/
Apache License 2.0
6.47k
stars
698
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
微调模型时train和eval的loss和梯度有时是nan
#677
ScottishFold007
opened
17 hours ago
0
请问一下这个shuffle_size参数的取值逻辑是什么，修改后会影响效果吗
#676
shenlou11
opened
20 hours ago
1
【VC模式】VC模式下的流式推理，已经是分段返回，但合成音频时仍然只有最后一段
#675
wang-TJ-20
closed
16 hours ago
2
怪异的表现
#674
jacksonjack001
opened
1 day ago
1
Question
#673
dillfrescott
closed
1 day ago
2
是否有支持mps的计划呀，mac上用cpu跑挺亏的
#672
ondineyuga
opened
3 days ago
1
Using an Instruct model for inference without embeddings, how can the speaker be specified?
#671
0xCAFEBABE0
opened
3 days ago
4
fix bug
#670
aluminumbox
closed
3 days ago
0
[Question]How to support batch inference?
#669
wjj19950828
opened
4 days ago
2
请教一下TransformerEncoderLayer和ConformerEncoderLayer类中的forward函数的一些问题， 这个函数中的pos_emb的维度是怎么确定的
#668
goingHan
opened
4 days ago
0
0样本声音克隆,部署后台进程,显存持续增加未自动释放
#667
abo123456789
opened
4 days ago
1
cosyvoice在克隆语音时出现问题了， Error: The size of tensor a (5002) must match the size of tensor b (2) at non-singleton dimension 3
#666
YuChaoM
opened
4 days ago
3
有支持8k采样率的模型吗？
#665
Dollhan
opened
4 days ago
1
Two times of fade-in-out
#664
OswaldoBornemann
opened
4 days ago
1
念古诗的时候断句不是根据标点而是根据词意
#663
interfacekun
opened
5 days ago
1
same loss when running two experiments simultaneously under examples/libritts/cosyvoice
#662
dbkest
opened
5 days ago
1
Unable to play generated audio in Safari
#661
klx1204
opened
6 days ago
2
关于300M-instruct模型的问题
#660
yyliuCecilia
opened
6 days ago
1
About "base continue train".
#659
sunnnnnnnny
opened
6 days ago
2
webui是否支持多路并发？开两个页面同时流式合成报错
#658
yangpeng-space
opened
6 days ago
1
WER计算
#657
liuxin99
opened
1 week ago
1
无法打开网页，会报错，好像和activations.py有关
#656
ztxxkaty
opened
1 week ago
2
【instruct模式性别变化】instruct模式下删除音色的embedding的原因
#655
wang-TJ-20
opened
1 week ago
1
流式模式下语音更完美的朗读
#654
lucasjinreal
opened
1 week ago
5
resume training
#653
aluminumbox
closed
1 week ago
0
从头训模型总是loss突然变成nan
#652
lilinqin
opened
1 week ago
1
如果想添加一些方言克隆，是需要重新训练模型吗？
#651
hjj-lmx
opened
1 week ago
3
25HZ的效果比50的差不少，会多字
#650
hjj-lmx
opened
1 week ago
1
请问如何提升复制音色的相似度
#649
linweijiang
closed
1 week ago
1
Is the SenseVoice-Large model currently commercialized?
#648
JinYuanZhang999
opened
1 week ago
1
No module named 'conformer'
#647
yyliuCecilia
opened
1 week ago
1
发现预定义音色在生成长音频时声音粗细、音色不稳定，而使用3s复制生成、不稳定问题会减轻。请问如何自定义音色，例如将3s复制生成的音色保存下来供使用
#646
yangpeng-space
opened
1 week ago
3
单3090TI训练爆内存
#645
ScottishFold007
opened
1 week ago
1
开头语音重复，末尾丢失
#644
ww18735135443
opened
1 week ago
1
Exporting the operator 'aten::scaled_dot_product_attention' to ONNX opset version 18 is not supported.
#643
donstang
opened
1 week ago
1
flow模型SFT训练不收敛
#642
JohnHerry
opened
1 week ago
3
“闵行（hang） 区”读成 “闵行（xing）”，这种多音字读错的问题有朋友遇到过吗？如何解决？
#641
timadidas
opened
1 week ago
1
关于 Speaker Interpolation 如何实现
#640
LongMarch7
opened
1 week ago
4
use stream read to save memory
#639
aluminumbox
closed
2 weeks ago
0
Streaming inference
#638
OswaldoBornemann
opened
2 weeks ago
1
最新版本 文本正则 丢字，官方例子下
#637
wantt
closed
1 week ago
1
data_pipeline中sort是必要的处理步骤吗
#636
shenlou11
opened
2 weeks ago
1
Update Dockerfile with CUDA 11.8
#635
swang109
opened
2 weeks ago
1
Ubuntu 运行 安装CosyVoice-ttsfrd报错'CosyVoiceFrontEnd' object has no attribute 'zh_tn_model' or 'en_tn_model'
#634
junge1010
opened
2 weeks ago
1
Flow 模型finetune 会逐渐崩掉?
#633
dyyoungg
closed
2 weeks ago
4
load_onnx=True比load_onnx=False 情况下慢
#632
goingHan
opened
2 weeks ago
1
流式合成有电音输出，如何解决
#631
doudou0601
opened
2 weeks ago
5
Why do length regulator in flow split token into three parts (head/mid/tail) ?
#630
zhangyike
opened
2 weeks ago
2
300M基础模型的断句错误问题
#629
czydfj
opened
2 weeks ago
2
flow中的attn_mask数据类型有问题
#628
NiHaoUCAS
opened
2 weeks ago
4
Next