adamnsandle
bfdc019302
add v6.2 model
2025-11-06 07:23:43 +00:00
Purfview
556a442942
Fix type hint for min_silence_at_max_speech (float -> int)
2025-11-04 08:30:01 +00:00
Dimitrii Voronin
9623ce72da
Merge pull request #710 from Purfview/patch-3
...
Fixes and refines - use_max_poss_sil_at_max_speech arg
2025-10-29 12:36:58 +03:00
adamnsandle
d8f88c9157
drop_chunks fix
2025-10-29 09:14:45 +00:00
Purfview
b15a216b47
Reword a comment
2025-10-24 10:30:34 +01:00
Purfview
2389039408
Fixes and refines - use_max_poss_sil_at_max_speech arg
...
Removed redundant "if temp_end != 0:" check.
Multiple "window_size_samples * i" - assigned to a variable.
Restored the previous functionality (which was broken) when use_max_poss_sil_at_max_speech=False.
@shashank14k was your https://github.com/snakers4/silero-vad/pull/664 PR still WIP when it was merged?
Anyway, please test if use_max_poss_sil_at_max_speech=True behaviour is same, and "False" is same as before your PR.
2025-10-24 07:46:41 +01:00
Purfview
81e8a48e25
Removes redundant hop_size_samples variable
...
Remove redundant hop_size_samples variable
2025-10-23 05:23:18 +01:00
Purfview
a30b5843c1
Fixes various typos
2025-10-23 04:02:13 +01:00
adamnsandle
77c91a91fa
resolve torchaudio 2.9 utils
2025-10-17 12:35:40 +00:00
Manuel Pariente
f1175c902f
Remove ipdb and raise error directly
2025-10-10 10:46:44 +02:00
adamnsandle
11631356a2
get rid of hop_size_ratio
2025-08-26 06:53:53 +00:00
Dimitrii Voronin
34dea51680
Merge pull request #664 from shashank14k/master
...
Adding additional params to get_speech_timestamps
2025-08-26 09:50:44 +03:00
adamnsandle
4c00cd14be
add v6 models
2025-08-25 14:56:50 +00:00
adamnsandle
1da76acfc3
fx
2025-08-25 14:07:32 +00:00
shashank14k
bbf22a0064
Added params for hop_size, and min_silence_at_max speech to cut at a possible silence when max_dur reached to avoid abrupt cuts
2025-07-25 20:51:40 +05:30
David Rust-Smith
22a2362b4c
Surface drop_chunks in init
2025-06-10 11:36:10 -07:00
Dimitrii Voronin
0dd45f0bcd
Merge pull request #626 from b3by/feature/process_chunks_in_seconds
...
Use second coordinates for audio concatenation in collect_chunks and drop_chunks
2025-03-24 19:02:56 +03:00
Antonio Bevilacqua
6622e562e4
time resolution can be specified when coordinates are returned in seconds
2025-03-24 08:53:28 +01:00
Antonio Bevilacqua
d5625d5c38
added audio concatenation for collect_chunks and drop_chunks based on second coordinates
2025-03-21 13:06:59 +01:00
adamnsandle
eaf633ec9d
fx negative ths bug
2024-11-25 13:54:46 +00:00
adamnsandle
782e30d28f
fx https://github.com/snakers4/silero-vad/issues/576
2024-11-22 08:17:25 +00:00
adamnsandle
d2ab7c254e
add just 16k model
2024-11-13 08:53:27 +00:00
adamnsandle
6217b08bbb
add other opsets
2024-11-12 08:25:06 +00:00
adamnsandle
e680ea6633
add half onnx model
2024-08-22 14:30:13 +00:00
adamnsandle
4109b107c1
add neg_threshold parameter explicitly
2024-08-20 08:53:15 +00:00
adamnsandle
dd0b143803
VadIterator first chunk bag fx
2024-07-15 10:37:46 +00:00
adamnsandle
f0d880d79c
make package structure
2024-07-09 08:26:17 +00:00
