The XMU Speech Lab (XMUSPEECH) of Xiamen University (XMU) currently includes three professors and more than twenty graduate students. The research directions of XMUSPEECH focus on environment (noise, channel) robust speech signal processing, speaker recognition and speech recognition. Since 2005, XMUSPEECH has conducted more than ten national-funded projects, obtained twenty patents and published more than thirty journal papers in SCI journals. Specially, environmental robust speech signal processing technologies of XMUSPEECH have found extensive practical applications, including intelligent robot, intelligent mobile and telecommunication. The developed embedded speech recognition chip has been applied into more than one hundred thousand robot toys. And the speaker recognition mobile software has also been downloaded more than one hundred thousand times.