This paper describes a number of experiments in adding new grammatical knowledge to the Berkeley Restaurant Project (BeRP), our medium-vocabulary (1300 word), speaker-independent, spontaneous continuous-speech understanding system (Jurafsky et al. 1994). We describe an algorithm for using a probabilistic Earley parser and a stochastic contextfree grammar (SCFG) to generate word transition probabilities at each frame for a Viterbi decoder. We show that using an SCFG as a language model improves word error rate from 34.6% (bigram) to 29.6% (SCFG), and semantic sentence recognition error from from 39.0% (bigram) to 34.1% (SCFG). In addition, we get a further reduction to 28.8%word error by mixing the bigram and SCFG LMs. We also report on our preliminary results from using discoursecontext information in the LM. 1. TIGHT COUPLING A number of researchers have proposed ways to use naturallanguage -backend information in the speech recognition process. Moore et al. (1989) used a unification...