想识别为拼音,但WER一直为1正常吗? #593
Unanswered
gezimonkey
asked this question in
Q&A
Replies: 1 comment
-
可以看下train的结果,如果还是unk的话再看vocab有没有问题。 |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
我尝试使用DeepSpeech训练自己的数据来识别带声调的拼音(例如:pu3 tong1 hua4)
数据大约有500小时采样率为16K的wav,质量一般,我进行了降噪和vad处理
训练时设置力lang_model_path为空,其他参数未动
epoch>60后,loss在0.6-1.2之间震荡
test结果全部为< u n k >
wer等于1
我想知道问题出在哪?
Beta Was this translation helpful? Give feedback.
All reactions