hamuchiwa
hamuchiwa
@BingLingGroup 谢谢回复，这么说数据保护什么的想也不要想了。。。还有一事不要明白，谷歌的这个[语音转文本](https://cloud.google.com/speech-to-text/)不是收费的嘛，作者找的这个api key还能无限量的用？
@BingLingGroup 请问有没有其他的语音转中文字幕的工具推荐，我用这个autosub转了一个视频，有些字幕不是很准确。
@BingLingGroup 文字内容大体正确，个别词语识别有误。我大概也上网搜了搜， 免费的除了这个也没找着别的，其他的都是收费的。有一个叫“人人译世界”的应用，可以免费试用生成一分钟的字幕，用的同样一个中文视频，100%识别正确，就是速度很慢。Youtube好像还不能识别中文吧。 我觉得准确率很重要，数据的保护也很重要啊，要是有能线下转字幕的就好了，曲线救国，多绕几个圈我也愿意~~
@Sirajabudabe @PuriHa If the opencv model doesn't work for you, try to create a CNN in Tensorflow/Keras. I have included an example in #173 in case you haven't tried yet.
@jhadkaterji https://github.com/hamuchiwa/AutoRCCar/issues/173#issuecomment-478619182
Hi, I use Mac, Python 2.7 and PyCharm.
I need more info, what errors are you seeing? Are you using python 2? Does the IP address and port on raspberry pi match the ones on the server? Does...
Hi, yashid: depends on your track, in my case it's about 300 images myxingchao: test files just added
MihaHoo: It works for me. I'm using Mac but I don't think there should be any difference. When you hit enter, the command line window should be grayed out, then...
MihaHoo: have you tried using a python IDE, for example, PyCharm?