|
|
|
|
|
|
|
|
|
convolve(signal,
kernel)
Returns something the length of the signal, by zero padding. |
source code
|
|
|
vowel(width,
ectrs,
neural,
neural_now,
Nsv,
formant_low=None,
formant_high=None) |
source code
|
|
|
spectral_entropy(width,
ectrs,
neural,
neural_now,
Nsv)
This is inspired (but only loosely) by "Robust Entropy-based
Endpoint Detection for Speech Recognition in Noisy
Environments." by Jia-lin Shen and Jeih-weih Hung and Lin-shan
Lee, http://www.ee.columbia.edu/~dpwe/papers/ShenHL98-endpoint.pdf
International Conference on Spoken Language Processing, 1998. |
source code
|
|
|
space_time_entropy(width,
ectrs,
neural,
neural_now,
Nsv) |
source code
|
|
|
haspitch(width,
ectrs,
neural,
neural_now,
Nsv) |
source code
|
|
|
peakiness(width,
ectrs,
neural,
neural_now,
Nsv) |
source code
|
|
|
pdur(ectrs,
neural,
neural_now,
Dt,
Cfac=1.0) |
source code
|
|
|
roughness(width,
ectrs,
neural,
neural_now,
Nsv) |
source code
|
|
|
dissonance(width,
ectrs,
neural,
neural_now,
Nsv) |
source code
|
|
|
vowel_edge(width,
ectrs,
neural,
neural_now,
Nsv,
do_abs=False,
formant_low=None,
formant_high=None) |
source code
|
|
|
fricative(width,
ectrs,
neural,
neural_now,
Nsv) |
source code
|
|
|
|
|
fricative_edge(width,
ectrs,
neural,
neural_now,
Nsv,
do_abs=False,
version=1) |
source code
|
|
|
|
|
normalize_neural(data,
dt,
Dt,
bmin=None,
bmax=None,
db=0.707106781187,
do_mod=0,
do_dissonance=False,
PlompBouman=False,
cache_info=None,
norm_kernel=None,
Nsv=None,
t_kernel=None,
f_kernel=None) |
source code
|
|