help_message.text

这是一个从音频信号里提取特征参量的程序

L"name: isip_recognize\nsynopsis: isip_recognize -param param.sof [options] [file1.sof...]\ndescription: speech recognizer, transforms speech to text\nexample: isip_recognize -param param.sof -list identifier_list.sof\n\noptions:\n -accum_file: accumulator file\n -accum_list: accumulator list\n -debug_level: debugging information level (default = NONE, BRIEF, DETAILED, ALL)\n -extension: output file extension\n -list: file containing input file list in Sdb format\n -log_file: log file used instead of stdout\n -dt_file the decision tree file\n -qa_file the question and answer file used for decision tree\n -merge_threshold the merge threshold for decision trees\n -split_threshold the split threshold for decision trees\n -occ_threshold the occupancy threshold for decision trees\n -output: directory to place output (default = source file directory)\n -output_file hypotheses.out: output file name in FILE mode\n -output_level the range of output level\n -update_level the range of update level\n -output_format: output file format (NIST_SCORE, SOF)\n -output_list out_list.sof: output files list in LIST mode\n -output_mode: output mode specification (FILE, TRANSFORM, LIST)\n -parameter_file: the parameter file\n -preserve: number of levels of subdirectories to preserve (default = 0)\n -verify: run without actual decoding\n -stream: process input from stdin rather than file\n -base: identifier for hypotheses. Used only if -stream is specified, and no\n\t  input file list is provided.\n -verbose: log verbosity level (default = NONE, BRIEF, DETAILED, ALL)\n\narguments: filname(s) and/or file list(s)\n\nman page: $ISIP/util/speech/isip_recognize/index.html"