连续语音识别Android的 - 无间隙 [英] Continuous Speech Recognition Android - Without Gaps
问题描述
我有一个活动,实现 RecognitionListener
。为了使它持续,每次 onEndOfSpeech()
我再次启动监听器:
I have an activity that implements RecognitionListener
. To make it continuous, every time onEndOfSpeech()
I start the listener again:
speech.startListening(recognizerIntent);
但是,这需要一定的时间(约半秒),直到它开始,所以有这样的一半的第二间隙,其中,没有被监听。因此,我想在这时间差被说的话。
But, it takes some time (around half a second) till it starts, so there is this half a second gap, where nothing is listening. Therefore, I miss words that were spoken in that time difference.
在另一方面,当我使用谷歌的语音输入,以决定消息而不是键盘 - 这个时间间隔不存在。含义 - 有一个解决方案。
On the other hand, when I use Google's Voice input, to dictate messages instead of the keyboard - this time gap does not exist. Meaning - there is a solution.
这是什么?
感谢
推荐答案
尝试寻找一些其他的API ....
try looking at a couple other api's....
演讲演示:具有源<一个href=\"https://github.com/scheib/chromium/blob/master/content/browser/speech/google_streaming_remote_engine.cc\"相对=nofollow>这里并讨论的此处和CLI 这里
speech demo : has source here and is discussed here and operated on CLI here
您可以使用全双工谷歌API(其速率上限为每天50)
you could use the full duplex google api ( its rate capped at 50 per day )
或者,如果你喜欢这个总体思路检查IBM的沃森讨论<一个href=\"https://developer.ibm.com/answers/questions/175476/watson-speech-to-text-response-not-continuous.html\"相对=nofollow>这里
Or if you like that general idea check ibm's watson discussed here
IMO - 其更复杂,但不封顶。
IMO - its more complex but not capped .
这篇关于连续语音识别Android的 - 无间隙的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!