http://cslt.org/mediawiki/index.php?action=history&feed=atom&title=2014-05-09 2026-04-15T07:34:55Z 本wiki的该页面的版本历史 MediaWiki 1.23.3 http://cslt.org/mediawiki/index.php?title=2014-05-09&diff=9871&oldid=prev 2014-05-09T01:45:57Z

<p>以内容“==Resoruce Building== * Maxi onboard * Release management should be started: Zhiyong (+) * Blaster 0.1 &amp; vivian 0.0 system release == Leftover questions== * Asymmetric...”创建新页面</p> <p><b>新页面</b></p><div>==Resoruce Building==<br /> * Maxi onboard<br /> * Release management should be started: Zhiyong (+)<br /> * Blaster 0.1 &amp; vivian 0.0 system release<br /> <br /> == Leftover questions==<br /> * Asymmetric window: Great improvement on training set(WER 34% to 24%), however the improvement is lost on test. Overfitting? <br /> * Multi GPU training: Error encountered<br /> * Multilanguage training<br /> * Investigating LOUDS FST. <br /> * CLG embedded decoder plus online compiler.<br /> * DNN-GMM co-training<br /> <br /> == AM development ==<br /> <br /> === Sparse DNN ===<br /> * GA-based block sparsity (++)<br /> :* Found a paper in 2000 with similar ideas. <br /> :* Try to get a student working on high performance computing to do the optimization <br /> <br /> ===Noise training===<br /> :* With-clean training done. Much better on clean testing<br /> :* Experiments done. Prepare paper. <br /> <br /> ===GFbank===<br /> <br /> * GFBank sinovoice 1400 MPE stream<br /> * GFBank sinovoice 6000 MPE stream<br /> <br /> ===Multilingual ASR===<br /> <br /> * MPE-based training is not very sensitive to data imbalance for English &amp; Chinese<br /> * Data duplication can trade-off the performance of two languages<br /> * Test sharing shemes<br /> <br /> ===Denoising &amp; Farfield ASR===<br /> <br /> * Baseline: close-talk model decode far-field speech: 92.65<br /> * Will investigate DAE model.<br /> <br /> ===VAD===<br /> <br /> * VAD bug fixed???<br /> * Test frame VAD accuracy<br /> <br /> ===Scoring===<br /> * Phone-sequence based graph decoding done<br /> * online scoring on going<br /> <br /> ==Word to Vector==<br /> * Paper writing<br /> <br /> ==LM development==<br /> <br /> ===NN LM===<br /> <br /> * Character-based NNLM (6700 chars, 7gram), 500M data training done.<br /> :* Inconsistent pattern in WER were found on Tenent test sets<br /> :* probably need to use another test set to do investigation. <br /> <br /> * Investigate MS RNN LM training<br /> <br /> ==QA==<br /> <br /> ===FST-based matching===<br /> :* Word-based FST 1-2 seconds with 1600 patterns. Huilan's implementation &lt;1 second.<br /> :* THRAX toolkit for grammar to FST<br /> <br /> * Investigate determinization of G embedding <br /> :* Refer to Kaldi new code</div>

Cslt