“Hulan-2013-12-27”版本间的差异
来自cslt Wiki
| 第10行： | 第10行： | ||
:* Deliver a female voice | :* Deliver a female voice | ||
:* Deliver a tunable pitch/speaking rate | :* Deliver a tunable pitch/speaking rate | ||
| − | :* Male voice | + | :* Male voice labeling finished |
:* Pronunciation dictionary check | :* Pronunciation dictionary check | ||
| 第28行： | 第28行： | ||
:* Extracting person names | :* Extracting person names | ||
:* Rule-based extraction, consider surname, name convention | :* Rule-based extraction, consider surname, name convention | ||
| − | :* N-gram | + | :* N-gram non-names filtering out |
| − | :* | + | :* Filter out candidates with active word embedded |
:* Check possible names by search engine | :* Check possible names by search engine | ||
| − | |||
=System design= | =System design= | ||
2013年12月27日 (五) 02:03的版本
目录
ASR
ASR Kernel development
TTS
- This week
- Deliver a female voice
- Deliver a tunable pitch/speaking rate
- Male voice labeling finished
- Pronunciation dictionary check
Dialog system
Statistical approach
- A preliminary book-search demo is ready, with 400 books, 6 properties, 10 patterns.
- Looks fine, while the pattern set is still limited
- Need to expand the patterns and properties and enrich the functionality.
- Looking for more efficient pattern match/learning approach.
Template matching
- New words extraction
- Extracting person names
- Rule-based extraction, consider surname, name convention
- N-gram non-names filtering out
- Filter out candidates with active word embedded
- Check possible names by search engine
System design
- Interface implementation on going (dimension management).
- Deadline is 14/1/17
