<speechrecog>

swirec_acoustic_adapt_adapt_model

Boolean

Suspends all self-learning for a given language (to save CPU).

swirec_acoustic_adapt_root

String

Specifies the default location for storing acoustic data.

swirec_acoustic_adapt_suppress_adaptation

Boolean

Suspends self-learning temporarily, or within a limited scope.

swirec_default_language

String

Specifies a default language for built-in grammars.

swirec_diag_tags_enable

String

Lists tags that are to be activated for diagnostic logging.

swirec_language_versions

String

Specifies which language versions to use when more than one is installed. See for more information and an example.

swirec_lmweight

Float

Weights language models relative to acoustic models.

swirec_load_adjusted_speedvsaccuracy

String

Balance processing speed relative to recognition accuracy.

swirec_max_arcs

Integer

Sets maximum levels for FSM (finite state machine) arcs.

swirec_model_name

String

Specifies the location of a custom recognition model to use.

swirec_phoneme_lookahead_beam

Float

Guides a Viterbi beam search.

swirec_sensitive_query_keys

String

Suppresses logging of specified key values.

swirec_state_beam

Float

Sets the main Viterbi beam search parameter(s).