SELMA: A Speech-Enabled Language Model for Virtual Assistant Interactions | IEEE Conference Publication | IEEE Xplore