Timeshrinking
From SpeechWiki
Fisher experiments
train <math>\tau</math> | test <math>\tau</math> | dev 2k utt WER | frames dropped |
---|---|---|---|
1 (old baseline) | 1 | 51.6% | 0% |
.6 | .6 | 69.3% | ~35% |
.9 | .9 | 56.3% | ~5% |
1 | .9 | 53.9% | ~5% |
.9 | 1 | TE | ~5% |
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.
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.