Hi, Does the AVS expects keyword/wake-up word in the start of the audio stream?( As per our understanding Amazon echo start audio streaming just before the keyword) If AVS expects keyword, will it remove the first word of audio stream in every request? or keyword is not mandatory for AVS? Does AVS supports any keyword/wake-up word or only "Alexa", because we want to use some other keyword (like wallnut/maya etc) for our development and testing.purpose. Thanks
Currently, AVS does not support any wake-word. AVS only enters the picture after you have determined that it is the user's intent to speak to Alexa, either by button press or your own far-field implementation. Additionally, if you implement your own far-field engine and would like to be certified for AVS, you will need to use the Alexa keyword.