Ticket #62 (new defect) — at Initial Version

Opened 16 years ago

Last modified 7 years ago

Warning: IW-triphone for word head "sp-ow+w" not found, fallback to pseudo {ow+w}

Reported by: kmaclean Owned by: kmaclean
Priority: major Milestone: Unassigned
Component: Speech Rec Engine Version: 0.1-alpha
Keywords: julian julius Cc:

Description

getting this error when running julius in batch mode:

/usr/local/julius/julius-3.5-linuxbin/bin/julian-3.5-std -input rawfile -filelist wavlst  -C julian.jconf -sscalc  > juliusOutput
}}

julian.jconf file

{{{
# Copyright (c) 1991-2005 Kawahara Lab., Kyoto University
# Copyright (c) 2000-2005 Shikano Lab., Nara Institute of Science and Technology
# Copyright (c) 2005      Julius project team, Nagoya Institute of Technology
# All rights reserved
#
# Example of Runtime Configuration File for Julian rev.3.5
#
# 1) NOTICE: relative paths must be relative to THIS FILE.
#	     Current working directory does not affect.
# 2) Blank lines and comments after '#' are ignored.
#    Use `\#' to specify `#' in an argument, 
# 3) Each line should be shorter than 512 bytes.
# 4) Syntax of each arguments are same as corresponding command options.
# 5) The commented-out values in this file are default values in Julian.
#

######################################################################
#### Files
######################################################################
##
## Grammar definition file (DFA and dictionary)
##

#### There are three ways to specify the grammar files.
#### (1) and (2) can be used multiple times.

#### (1) Specify by common prefix of .dfa and .dict files. Comma-separated
#### prefixes can be specified for multiple grammar recognition
#-gram /cdrom/testrun/sample_grammars/vfr/vfr

#### (2) Or you can give Julian a text file which contains list of grammar
#### prefixes one per line.
#-gramlist file

#### (3) Classic way to specify a grammar.
-dfa sample.dfa
-v sample.dict

#### If you want to clear previously specified grammars, use this at the
#### point.
#-nogram

##
## Acoustic HMM file
##
# support ascii hmmdefs or binary format (converted by "mkbinhmm")
# format (ascii/binary) will be automatically detected
-h hmmdefs

## triphone model needs HMMList that maps logical triphone to physical ones.
-hlist tiedlist

######################################################################
#### Multiple grammar recognition
######################################################################
#-multigramout		# Output results for each grammar

######################################################################
#### Language Model 
######################################################################
##
## word insertion penalty
##
#-penalty1 0.0		# first pass
#-penalty2 0.0		# second pass

######################################################################
#### Dictionary
######################################################################
##
## do not giveup startup on error words
##
#-forcedict

######################################################################
#### Acoustic Model
######################################################################
##
## Context-dependency handling will be enabled according to the model type.
## Try below if julius wrongly detect the type of hmmdefs
##
#-no_ccd		# disable context-dependency handling
#-force_ccd		# enable context-dependency handling

## 
## If julius go wrong with checking parameter type, try below.
##
#-notypecheck
#

##
## (PTM/triphone) switch computation method of IWCD on 1st pass
##
#-iwcd1 best N	# assign average of N-best likelihood of the same context
-iwcd1 max	# assign maximum likelihood of the same context
#-iwcd1 avg	# assign average likelihood of the same context (default)

######################################################################
#### Gaussian Pruning
######################################################################
## Number of mixtures to select in a mixture pdf.
## This default value is optimized for IPA99's PTM,
## with 64 Gaussians per codebook
#-tmix 2

## Select Gaussian pruning algorithm
## defulat: beam (standard setting), safe (others)
-gprune safe		# safe pruning, accurate but slow
#-gprune heuristic	# heuristic pruning
#-gprune beam		# beam pruning, fast but sensitive
#-gprune none		# no pruning

######################################################################
#### Gaussian Mixture Selection 
######################################################################
#-gshmm hmmdefs		# monophone HMM for GMS
			# (OFF when not specified)
#-gsnum 24		# number of states to be selected on GMS

######################################################################
#### Search Parameters
######################################################################
#-b 2000                 # beam width on 1st pass (#nodes) for monophone
#-b 800                 # beam width on 1st pass (#nodes) for triphone,PTM
#-b 1000                # beam width on 1st pass (#nodes) for triphone,PTM,engine=v2.1
#-b2 100                 # beam width on 2nd pass (#words)
#-sb 80.0		# score beam envelope threshold
#-s 500                 # hypotheses stack size on 2nd pass (#hypo)
#-m 2000                # hypotheses overflow threshold (#hypo)
#-lookuprange 5         # lookup range for word expansion (#frame)
#-n 1                   # num of sentences to find (#sentence)
#-n 10                  #   (default for 'standard' configuration)
#-output 1              # num of found sentences to output (#sentence)
#-looktrellis		# search within only backtrellis words

######################################################################
#### Inter-word Short Pause Handling
######################################################################
##
## Specify short pause model name to be treated as special
##
#-spmodel "sp"		# HMM model name

##
## For insertion of context-free short-term inter-word pauses between words
##  (multi-path version only)
##
#-iwsp			# append a skippable sp model at all word ends
#-iwsppenalty 0.0	# transition penalty for the appenede sp models

######################################################################
#### Speech Input Source
######################################################################
## select one (default: mfcfile)
#-input mfcfile         # MFCC file in HTK parameter file format
#-input rawfile         # raw wavefile (auto-detect format)
                        # WAV(16bit) or
                        # RAW(16bit(signed short),mono,big-endian)
                        # AIFF,AU (with libsndfile extension)
			# other than 16kHz, sampling rate should be specified
			# by "-smpFreq" option
#-input mic             # direct microphone input
			# device name can be specified via env. val. "AUDIODEV"
#-input netaudio -NA host:0    # direct input from DatLink(NetAudio) host
#-input adinnet -adport portnum # via adinnet network client
#-input stdin		# from standard tty input (pipe)

#-filelist filename	# specify file list to be recognized in batch mode

#-nostrip		# switch OFF dropping of invalid input segment.
                        # (default: strip off invalid segment (0 sequence etc.)
#-zmean			# enable DC offset removal (invalid for mfcfile input)

######################################################################
#### Recording
######################################################################
#-record directory	# auto-save recognized speech data into the dir

######################################################################
#### GMM-based Input Verification and Rejection
######################################################################
#-gmm gmmdefs		# specify GMM definition file in HTK format
#-gmmnum 10		# num of Gaussians to be computed per mixture
#-gmmreject "noise,laugh,cough" # list of GMM names to be rejected

######################################################################
#### Too Short Input Rejection
######################################################################
-rejectshort 2000	# reject input shorter than specified millisecond

######################################################################
#### Speech Detection
######################################################################
#-pausesegment		# turn on speech detection by level and zero-cross
#-nopausesegment	# turn off speech detection by level and zero-cross
			# (default: on for mic or adinnet, off for file)
#-lv 2000		# threshold of input level (0-32767)
#-headmargin 300	# head margin of input segment (msec)
#-tailmargin 400	# tail margin of input segment (msec)
#-zc 60			# threshold of number of zero-cross in a second

######################################################################
#### Acoustic Analysis
######################################################################
-smpFreq 8000		# sampling rate (Hz)
#-smpPeriod 625		# sampling period (ns) (= 10000000 / smpFreq)
#-fsize 400		# window size (samples)
#-fshift 160		# frame shift (samples)
#-delwin 2		# delta window (frames)
#-hifreq -1		# cut-off hi frequency (Hz) (-1: disable)
#-lofreq -1		# cut-off low frequency (Hz) (-1: disable)
#-cmnsave filename	# save CMN param to file (update per input)
#-cmnload filename	# load initial CMN param from file on startup

######################################################################
#### Spectral Subtraction (SS)
######################################################################
#-sscalc		# do SS using head silence (file input only)
#-sscalclen 300		# length of head silence for SS (msec)
#-ssload filename       # load constant noise spectrum from file for SS
#-ssalpha 2.0		# alpha coef. for SS
#-ssfloor 0.5		# spectral floor for SS

######################################################################
#### Forced alignment
######################################################################
#-walign		# do forced alignment with result per word 
#-palign		# do forced alignment with result per phoneme
#-salign		# do forced alignment with result per HMM state

######################################################################
#### Word Confidence Scoring
######################################################################
#-cmalpha 0.05		# smoothing coef. alpha

######################################################################
#### Output 
######################################################################
#-separatescore		# output language and acoustic score separately
#-progout		# output partial result per a time interval
#-proginterval 300	# time interval for "-progout" (msec)
#-quiet			# output minimal result
#-demo			# = "-progout -quiet", suitable for dictation demo
#-debug			# output full message for debug
#-charconv from to	# output character set conversion (see manual for
			# available code set name)

######################################################################
#### Server module mode
######################################################################
#-module		# Run Julius on "Server module mode"
#-module 5530		# (when using another port number for connection)
#-outcode WLPSC		# select output message toward module (WLPSCwlps)

######################################################################
#### Misc.
######################################################################
#-help			# output help and exit
#-setting		# output engine configuration and exit
#-C jconffile		# expand other jconf file in its place

################################################################# end of file

}}}
Note: See TracTickets for help on using tickets.