ASR:2014-12-01
来自cslt Wiki
目录
Speech Processing
AM development
Environment
- Already buy 3 760GPU
- grid-9/12 760GPU crashed again; grid-11 shutdown automatically.
- Change 760gpu card of grid-12 and grid-14(+).
Sparse DNN
- Performance improvement found when pruned slightly
- need retraining for unpruned one; training loss
- details at http://liuc.cslt.org/pages/sparse.html
- HOLD
RNN AM
- Initial nnet seems not very well, need to be pre-trained or test lower learn-rate.
- For AURORA 4 1h/epoch, model train done.
- Using AURORA 4 short-sentence with a smaller number of targets.(+)
- Adjusting the learning rate.(+)
- Trying toolkit of Microsoft.(+)
- details at http://liuc.cslt.org/pages/rnnam.html
- Reading papers
A new nnet training scheduler
- Initial code done. No better than original one considering of taking much more iterations.
- details at http://liuc.cslt.org/pages/nnet-sched.html
Drop out & Rectification & convolutive network
- Drop out(+)
- AURORA4 dataset
- Use different proportion of noise data to investigate the effect of xEnt and mpe and dropout
- Problem 1) The effect of dropout in different noise proportion;
- Use different proportion of noise data to investigate the effect of xEnt and mpe and dropout
2) The effect of MPE in different noise proportion;
3) The effect of MPE+dropout in different noise proportion.
- Find and test unknown noise test-data.(++)
- Have done the droptout on normal trained XEnt NNET , eg wsj(learn-rate:1e-4/1e-5). Seems small learn-rate get the balance of accuracy and train-time.
- Debug the low cv frame-accuracy
- MaxOut(+)
- pretraining based maxout
- Select units in Groupsize interval, but need low learn-rate
- Force accept the first iteration. Jump out from the local-minimum
- pretraining based maxout
- SoftMaxout
- P-norm
- Need to solve the too small learning-rate problem
- Add one normalization layer after the pnorm-layer
- Need to solve the too small learning-rate problem
- Convolutive network (+)
- AURORA 4
-------------------------------------------------------------------------------------------------------------------------- nonlda | %WER |Dnn l-u | pool size-step| cnn dim-step-num | cnn_init_opts
cnn_std | 5.73 | 4 - 1200 | 3 - 3 | 8-1-128 512-128-256 |--patch-dim1 8
| | | | |--input_dim~patch-dim1
cnn_cnnunit_384 | 5.85 | 4 - 1200 | 3 - 3 | 8-1-128 512-128-384 |--patch-dim1 8
| | | | |--num-filters2 384
cnn_patchdim1_5 | 5.92 | 4 - 1200 | 3 - 3 | 5-1-128 512-128-256 |--patch-dim1 5
cnn_patchdim1_11 | 6.05 | 4 - 1200 | 3 - 3 | 11-1-128 512-128-256 |--patch-dim1 11
cnn_delta_1 | 5.98 | 4 - 1200 | 3 - 3 | 8-1-128 512-128-256 |--patch-dim1 8
cnn_delta_2 | 6.05 | 4 - 1200 | 3 - 3 | 8-1-128 512-128-256 |--patch-dim1 8
cnn_layer_3 | 6.00 | 4 - 1200 | 3 - 3 3 - 1 | 8-1-128 512-128-256 768-256-512 |--patch-dim1 8
cnn_layer_3_2 | 5.85 | 4 - 1200 | 3 - 3 2 - 2 | 8-1-128 512-128-256 768-256-512 |--patch-dim1 8
cnn_layer_3_3 | 5.73 | 4 - 1200 | 3 - 3 2 - 2 | 8-1-128 512-128-256 512-256-512 |--patch-dim1 8
cnn_layer_3_4 | 5.96 | 4 - 1200 | 3 - 3 2 - 2 | 8-1-128 512-128-256 256-256-512 |--patch-dim1 8
DAE(Deep Atuo-Encode)
(1) train_clean
drop-retention/testcase(WER)| test_clean_wv1 | test_airport_wv1 | test_babble_wv1 | test_car_wv1
---------------------------------------------------------------------------------------------------------
std-xEnt-sigmoid-baseline| 6.04 | 29.91 | 27.76 | 16.37
---------------------------------------------------------------------------------------------------------
std+dae_cmvn_noFT_2-1200 | 7.10 | 15.33 | 16.58 | 9.23
---------------------------------------------------------------------------------------------------------
std+dae_cmvn_splice5_2-100 | 8.19 | 15.21 | 15.25 | 9.31
---------------------------------------------------------------------------------------------------------
Denoising & Farfield ASR
- ICASSP paper submitted.
- HOLD
VAD
- Frame energy feature extraction, done
- Harmonics and Teager energy features being investigation (++)
- Previous results to be organized for a paper
- MPE model VAD ,good performance observed.
Speech rate training
- Data ready on tencent set; some errors on speech rate dependent model
- Retrain new model(+)
Scoring
- Timber Comparison done.
- harmonics based timber comparison: frequency based feature is better
- GMM based timber comparison is done. Similar to speaker recognition
- TODO: Code checkin and technique report
Confidence
- Reproduce the experiments on fisher dataset.
- Use the fisher DNN model to decode all-wsj dataset
- preparing scoring for puqiang data
Speaker ID
- Preparing GMM-based server.
- EER ~ 4% (GMM-based system)--Text independent
- EER ~ 6%(1s) / 0.5%(5s) (GMM-based system)--Text dependent
- test different number of components; fast i-vector computing
Language ID
- GMM-based language is ready.
- Delivered to Jietong
- Prepare the test-case
Voice Conversion
- Yiye is reading materials
Text Processing
LM development
Domain specific LM
- domain lm(need to discuss with xiaoxi)
- embedded language model(this week)
- train some more LMs with Zhenlong (dianzishu sogou bbs chosen)("need result").
- keep on training sogou2T lm(14/16 on 3rd iteration).(this week)
- new dict.
- handover of this work to hanzhenglong, give a simple docuemnt(this week)
tag LM
- different weight 2014-Nov-23,Monday
| method | tag-jsgf | corpus | weight | wer | ser | add_wer |
|---|---|---|---|---|---|---|
| experiment 3 | 500(490 less frequent and 10 unseen) | 500 | 0.1 | 16.72 | 77.92 | - |
| 0.3 | 15.42 | 71.25 | - | |||
| 0.5 | 15.40 | 69.58 | - | |||
| 0.7 | 15.28 | 68.75 | - | |||
| 0.8 | 15.38 | 68.33 | - | |||
| 1 | 15.98 | 69.17 | - | |||
| 2 | 19.08 | 70.83 | - | |||
| experiment 4 | 100(90 less frequent and 10 unseen) | 100 | 0.008 | 15.28 | 69.58 | - |
| 0.02 | 14.84 | 69.58 | - | |||
| 0.05 | 15.11 | 69.58 | - | |||
| 0.1 | 15.30 | 69.75 | - | |||
| 0.3 | 16.01 | 70.42 | - | |||
| experiment 5 | 500 | 100 | 0.01 | 17.57 | 78.75 | - |
| 0.05 | 16.84 | 77.08 | - | |||
| 0.08 | 16.59 | 76.25 | - | |||
| 0.15 | 16.76 | 75.42 | - | |||
| experiment 6 | 1280 | 500 | 0.1 | 17.42 | 77.92 | - |
| 0.5 | 15.20 | 69.17 | - | |||
| 0.8 | 15.30 | 68.33 | - | |||
| 1 | 15.69 | 69.58 | - |
- conclusion:
1. compare experiment 3 with experiment 5: same jsgf file, but the tag number in corpus if different, we can find that when add more tag to corpus, the optimal weight is larger. 2. compare experiment 3 with experiment 6: same tag number in corpus, but different jsgf size, we can find that different jsgf size have the same optimal weight.
- need to do
- tag Probability should test add the weight(hanzhenglong) and handover to hanzhenglong (this week)
- make a summary about tag-lm and journal paper(wxx and yuanb)(two weeks).
RNN LM
- rnn
- test wer RNNLM on Chinese data from jietong-data(this week)
- check the rnnlm code about how to Initialize and update learning rate.
- generate the ngram model from rnnlm and test the ppl with different size txt.(this week)
- lstm+rnn
- check the lstm-rnnlm code about how to Initialize and update learning rate.
Word2Vector
W2V based doc classification
- Initial results variable Bayesian GMM obtained. Performance is not as good as the conventional GMM.(hold)
- Non-linear inter-language transform: English-Spanish-Czch: wv model training done, transform model on investigation
Knowledge vector
- Knowledge vector started
- begin to code
Character to wordr
- Character to word conversion(hold)
- prepare the task: word similarity
- prepare the dict.
Translation
- v5.0 demo released
- cut the dict and use new segment-tool
QA
deatil:[1]
Spell mistake
- retrain the ngram model(caoli)
improve fuzzy match
- add Synonyms similarity using MERT-4 method(hold)
improve lucene search
- using MERT-4 method to get good value of multi-feature.like IDF,NER,baidu_weight,keyword etc.(liurong this month)
Multi-Scene Recognition
- handover to duxk(this week)
XiaoI framework
- give a report about xiaoI framework
- new inter will install SEMPRE
patent
- GA-method improve the QA(this week)
