search

modelscope / 3D-Speaker

A Repository for Single- and Multi-modal Speaker Verification, Speaker Recognition and Speaker Diarization
Apache License 2.0
1.2k stars 101 forks source link

训练问题 #124

Closed Onlyooo closed 3 months ago

Onlyooo commented 3 months ago

我想跑通这个CAM++模型的时候遇到了 1721024757886 请问这个怎么解决？

yfchenlucky commented 3 months ago

Wav文件没准备好，请从stage1开始运行并确认文件是否存在且正确。

Onlyooo commented 3 months ago

明白了谢谢您，还想请问如果我想做关于声音识别杂音这样的项目，应该通过哪个项目中的哪个网络比较好呢？

yfchenlucky commented 3 months ago

ERes2NetV2, CAM++, ECAPA-TDNN都可以试试。