Correction in State Duration Modeling for HMM-Based Speech Synthesis
In , ,
the probability of staying at state
given an observation sequence
is the probability of being in state
and we defined
and the variance
of the state duration density
is obtained as
the previous definition of
is statistially incorrect
because the state transitions were not taken into account.
in a statistically correct manner as
denotes the state at time
denotes the parmeter set of the HMM,
the forward and backward variables,
denote the state transition probability and the output probability,
- T. Yoshimura, K. Tokuda, T. Masuko, T. Kobayashi, and T. Kitamura,
``Simultaneous modeling of spectrum, pitch and duration
in HMM-based speech synthesis,''
IEICE Trans. D-II, vol.J83-D-II, no.11, pp.2099--2107, Nov. 2000
- T. Yoshimura, T. Masuko, K. Tokuda, T. Kobayashi, and T. Kitamura,
``Duration modeling for HMM-based speech synthesis,''
Proc. ICSLP-98, vol.2, Tu3A4, pp.29--32, Nov. 1998.