[Subject Prev][Subject Next][Thread Prev][Thread Next][Date Index][Thread Index]

[hts-users:00886] Re: Training by imposing decision trees


Simon King wrote (2007/10/27 15:11):

- use HHEd's "AU" command to synthesise *all* the models you need (the ones you previously deleted from the MMF, plus any unseen ones)

One other small wrinkle - the HHEd manual says about the AU command:

"If the clustering was performed at the state level a prototype model (an example of the same phone model occurring in a different context) is found and a new HMM is constructed that shares the transition matrix with the prototype model but consists of tied states selected using the decision tree."

so (since HTS uses state clustering) you would have to have some models loaded into HHEd in order to form these prototypes - I suppose you could have just one such model per phone, in a bogus context, and then strip out these models after you have synthesised the models you want

Simon


References
[hts-users:00876] Training by imposing decision trees, Thomas Drugman
[hts-users:00877] Re: Training by imposing decision trees, Heiga ZEN (Byung Ha CHUN)
[hts-users:00884] Re: Training by imposing decision trees, Simon King
[hts-users:00885] Re: Training by imposing decision trees, Heiga ZEN (Byung Ha CHUN)