The Alexa Skills Kit does not provide this. I think a few people have requested it as a feature. You might go and +1 those threads. Personally, I do not think that Amazon will do this. They put a huge amount of effort into sound processing and I can't really see anyone really improving upon this. And, it would also open up huge privacy concerns. If they control the filtering of the raw sound, they can limit a lot of the liability that an always-on/always-listening device has in a domestic context.
Thank for reply. Yeah, it is reasonable. however, if developer can access the audio data that transffered and used by alexa or alexa can provide some feature to provide after processing, it would be better for flexibility and extensibility. User or developer can do more with them. Thanks