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_segment_concat\nsynopsis: isip_segment_concat [options]\ndescr: generates minimum and maximum value for each dimension from the input feature-vectors. creates the segmental features given the ratio\nexample: isip_segment_concat -min_max -min_max_type text -min_max_file min_max.text -list identifiers_sdb.sof -input_format raw -input_type binary -dimension 39\nexample:isip_segment_concat -normalize -duration -min_max_file min_max.text -list identifiers_sdb.sof -audiodb audiodb.sof -input_format raw -input_type binary -dimension 39 -ratio 4:3:4 -transdb transdb.sof -level phone -output_type binary -output_format raw\n\noptions:\n -ratio: ratio used to concat the segmental features\n -min_max_type: output min_max file type(default=text, binary)\n -min_max_file: output min_max file\n -list input identifier list as sdb\n -audiodb: audio database specifying the input feature files\n -input_format: input feature file format(default=sof, raw)\n -input_type: input feature file type(default=binary,text)\n -output_format: output feature file format(default=sof, raw)\n -output_type: output feature file type(default=binary,text)\n -transdb: transcription database specifying the aligments\n -level transcription level for the alignments\n -debug: debug level (default=NONE, BRIEF, DETAILED, ALL)\n -help: this message\n -log_file: log messages\n\narguments:\n none\n\n flags:\n -min_max: generate min_max file\n -normalize: normalize input features before computing segmental features\n -duration: log-duration of the alignment length added to the output feature-vector\n\nman page: $ISIP/util/speech/isip_make_db/index.html"