On Mon, 28 Oct 2002, robert b wrote:
> > INFO: main.c(168): No lexical transcripts provided
> > INFO: corpus.c(1236): Will process all remaining utts starting at 0
> > INFO: main.c(272): Will produce FEAT dump
> > INFO: main.c(426): Writing frames to one file
> > INFO: agg_all_seg.c(149): 
> > FATAL_ERROR: "corpus.c", line 262: input string too long. Truncated.
> If by "removed the final linebreak", you mean that you went into Emacs and wrote out the file
> without a final newline, then that definitely won't work.
Ah well. So forgetting about the FATAL_ERROR, which I'm guessing I caused
by removing that linebreak, I'll keep thinking about that "No lexical
transcripts provided" error.
> Just shooting from the hip here: I didn't see any errors in your first run. Did you check all the
> log files? You might just do
> % grep -i error logdir/*/*
> I'll try doing this some time today.
Did that. That's where I got the errors that I reported -- the above INFO
lines were taken from logdir/01.vector_quantize/time.vq.agg_seg.log.
There was also a final error in time.log:
INFO: main.c(101): Reading
WARN: "s3io.c", line 253: Unable to open
model_parameters/time.cd_semi_6000_delinterp/mixture_weights for reading;
No such file or directory
but that just means that something broke earlier on and didn't write out
the s3 models, so that we then can't make the s2 models. So that's not a
very helpful error to look into more, I think.
"Users complain that they receive too much spam, while spammers protest
messages are legal." -InfoWorld
"You do not have to do everything disagreeable that you have a right to
do." -Judith Martin (Miss Manners)