5.3.4 CI Phone Models: Mixture Splitting
For technical details of mixture splitting, revisit Section 5.2.5. The mixture splitting process for context independent phone models is nearly identical to that of word models. This section provides instructions for splitting two, four, and eight mixtures and training the acoustic models after they're split. Follow the set of steps below: 2 Mixtures From the directory:
Command: isip_recognize -param params_2mix_split.sof -list $ISIP_TUTORIAL/research/isip/databases/lists/identifiers_train.sof -verbose all Version: 1.23 (not released) 2003/05/21 23:10:45 *** no audio database file was specified *** *** no symbol graph database file was specified *** *** no transcription database file was specified *** *** no front-end file was specified *** loading language model: $ISIP_TUTORIAL/models/ci_phone_models/lm_phone_jsgf_sp.sof loading statistical model pool: $ISIP_TUTORIAL/models/ci_phone_models/smp_phone_sp.sof *** no configuration file was specified *** ....Now, train the mixures with the command:
Command: isip_recognize -param params_2mix.sof -list $ISIP_TUTORIAL/research/isip/databases/lists/identifiers_train.sof -verbose all Version: 1.23 (not released) 2003/05/21 23:10:45 loading audio database: $ISIP_TUTORIAL/research/isip/databases/db/tidigits_audio_db.sof *** no symbol graph database file was specified *** loading transcription database: $ISIP_TUTORIAL/research/isip/databases/db/tidigits_trans_word_db.sof loading front-end: $ISIP_TUTORIAL/recipes/frontend.sof loading language model: $ISIP_TUTORIAL/models/ci_phone_models/lm_phone_jsgf_2mix_split.sof loading statistical model pool: $ISIP_TUTORIAL/models/ci_phone_models/smp_phone_2mix_split.sof loading configuration file: $ISIP_TUTORIAL/sections/s05/s05_02_p04/config.sof starting iteration: 0 processing file 1 (ae_12a): $ISIP_TUTORIAL/research/isip/databases/sof_8k/train/ae_12a.sof retrieving annotation graph for identifier: ae_12a, level: word transcription: ONE TWO average utterance probability: -71.081872936913655, number of frames: 110 processing file 2 (ae_1a): $ISIP_TUTORIAL/research/isip/databases/sof_8k/train/ae_1a.sof retrieving annotation graph for identifier: ae_1a, level: word transcription: ONE ....4 Mixtures Run the following command to split the mixtures:
Run the following command to split the mixtures:
|
