Timeshrinking

From SpeechWiki

Contents 1 Fisher experiments 2 Things to try 3 LM penalty and scale 4 Future Directions

Fisher experiments

number of frames dropped on Fisher corpus

<math>\tau</math>	frames dropped
1	0%
.9	~5%
.6	~35%

I have to check for bugs. It could be that the threshold is too low or it could be something else too. We should probably rerun baseline too, just to make sure I didn't optimize it unfairly.

Things to try

• Test svitchboard with fisher-trained model to see if we still get good results
• Train and test on plp+mlp, like svitchboard timeshrinking was done.
• Do baseline train+test to see if something changed in going from baseline to timeshrink structure files. (done, helped)

LM penalty and scale

Since we now have 62 PLP+MLP features instead of 39 PLP, we should probably change LM scale by a factor 62/39=1.58. The original (not carefully tuned)PLP LM scale was 10. Perhaps it would make sense to multiply the LM penalty (-1 for PLP) by the same 1.58 factor.

Future Directions

• Can be viewed as a two-mode special case of best-first viterbi search. So make a real best-first lattice search. Mark mentioned some attempts in the 80'ies to do this.

Timeshrinking results on fisher

train <math>\tau</math>	test <math>\tau</math>	dev 2000 utt WER	dev 2000 utt on triphone single-gaussian model WER	comments
1	1	51.6%		old baseline
1	1	53.7%	78.2%	baseline rerun exactly as timeshrinking to really make sure it's not getting an unfair advantage
.6	.6	69.3%
.9	.9	56.3%	80.4%
1	.9	53.9%
.9	.9	57.2	80.7%	using the non-timeshrinking str file for test
.9	1	54.6
1	1	55.4%	72.8%	PLP+MLP tandem
1	1	50.6%		PLP+MLP tandem, LM scale 16
.9	.9	TR	73.1%	PLP+MLP tandem