search

xmos / fwk_voice

Voice Framework
Other
11 stars 19 forks source link

Port py_vnr v0.2.1 changes to fwk_voice #336

Closed shuchitak closed 2 years ago

shuchitak commented 2 years ago

Results:

Current head of develop of Avona develop_results_Avona_prev_arch_xcore.csv develop_results_Avona_alt_arch_xcore.csv

With py_vnr v2.1.0 ported vnr_v2_1_0_results_Avona_prev_arch_xcore.csv vnr_v2_1_0_results_Avona_alt_arch_xcore.csv

With py_vnr v2.1.0 and the highpass filter (which is disabled by default) on MEL output enabled hpf_vnr_v2_1_0_results_Avona_prev_arch_xcore.csv hpf_vnr_v2_1_0_results_Avona_alt_arch_xcore.csv

shuchitak commented 2 years ago

py_vnr v2.1.0 vs head of Avona develop:

Prev-arch vnr_v2_1_0_prev_arch vnr_v2_1_0_prev_arch.csv

Alt-arch vnr_v2_1_0_alt_arch vnr_v2_1_0_alt_arch.csv

shuchitak commented 2 years ago

py_vnr v2.1.0, with hpf

Prev-arch hpf_vnr_v2_1_0_prev_arch hpf_vnr_v2_1_0_prev_arch.csv

Alt-arch hpf_vnr_v2_1_0_alt_arch hpf_vnr_v2_1_0_alt_arch.csv

andrewdewhurst commented 2 years ago

Agreed with Shuchita, Pavel, Alan and Thanasis that this is complete.