delayed-streams-modeling

rust/delayed-streams-modeling

Fork 0

mirror of https://github.com/kyutai-labs/delayed-streams-modeling.git synced 2025-12-22 19:09:57 +00:00

Commit Graph

Select branches

Hide Pull Requests

LaurentMazare-patch-1

bump-versions

changing_streaming

clippy-fix

configurable-api-key

contrib

faq

fix

fix-tts-stream

give_uv_instructions

local-voices

main

mlx-0.2.9

mlx-kv-cache-bug

mlx-mic

mlx-status

mlx-tts

mlx-tts.handle-stdin

mlx-tts.on_frahme

mlx-vad-example

precommit-ci

print-gen

pth-tts-streaming

ruff

streaming-inptu

tts-mlx-streaming

tts-pth-streaming

tts-pytorch.device

vad-python-example

vad-repos

vv/clean-up-examples

vv/clean-up-examples-2

vv/document-tts-config

vv/fix-example-usage

vv/fix-pre-commit

vv/mailing-list

vv/moshi-version

vv/readme

vv/rust-example

vv/rust-server-instructions

vv/shorter-names

vv/tokenizer-repo

vv/tts-pytorch

vv/tts-pytorch-nb

#1

#100

#102

#103

#107

#108

#114

#120

#122

#124

#125

#126

#13

#130

#142

#15

#153

#154

#16

#161

#162

#163

#164

#18

#19

#2

#21

#24

#26

#27

#28

#29

#3

#30

#31

#32

#33

#34

#35

#36

#37

#38

#39

#45

#55

#56

#6

#62

#63

#65

#66

#68

#7

#70

#82

#83

#84

#85

#88

#9

#91

13f343f88c Document TTS configuration better (#142) main Václav Volhejn 2025-11-26 19:19:28 +01:00
51fc312d7b Improve CFG documentation vv/document-tts-config Václav Volhejn 2025-11-20 10:56:59 +01:00
8fb54b2b07 Fix typo in README.md (#161) Bissmella Bahaduri 2025-11-20 10:15:45 +01:00
d5a3fb1705 Document TTS configuration better Vaclav Volhejn 2025-09-22 10:02:16 +02:00
f6074cb684 fix notebook for safetensors Alexandre Défossez 2025-09-11 17:22:30 +02:00
22e0b400e8 Replace "pre-print coming soon" Vaclav Volhejn 2025-09-11 13:00:03 +02:00
f741fc473f Add citation section to README Alexandre Défossez 2025-09-11 10:43:02 +02:00
263ed6ac04 Fix reading file encoding with UTF-8 (#130) Mohaidoss 2025-09-01 10:17:51 +02:00
ada17799b8 Bugfix. (#126) Laurent Mazare 2025-08-26 14:40:43 +02:00
ac7fc41d96 Bugfix. fix laurent 2025-08-26 14:38:20 +02:00
b331930f8e Refactor frame callback to remove last_time tracking (#125) Laurent Mazare 2025-08-26 14:36:27 +02:00
7fe21a179b Formatting. LaurentMazare-patch-1 laurent 2025-08-26 14:35:10 +02:00
1f1b43a1b4 Refactor frame callback to remove last_time tracking Laurent Mazare 2025-08-26 14:31:49 +02:00
2d301c9da0 Enable ruff in the pre-commit hooks (#124) Laurent Mazare 2025-08-26 13:48:11 +02:00
3c4ecdb4ab Install uv in the CI. ruff laurent 2025-08-26 13:00:41 +02:00
c5f933296a Disable the old hooks. laurent 2025-08-26 12:56:16 +02:00
2084eabd0c Enable ruff in the pre-commit hooks. laurent 2025-08-26 12:55:31 +02:00
f4016a8844 Clippy fix. (#122) Laurent Mazare 2025-08-25 09:33:07 +02:00
9c6b2675b3 Clippy fix. clippy-fix laurent 2025-08-25 09:30:48 +02:00
7d1e4b703a Add quantization support and GGUF loading to standalone STT Rust script (#120) Bai Li 2025-08-25 00:28:48 -07:00
affc0a052b Display the generated audio length in the mlx script. (#114) Laurent Mazare 2025-08-13 07:17:04 +02:00
7183c8ac6e Display the generated audio length in the mlx script. mlx-status Laurent 2025-08-13 07:16:08 +02:00
cf97f8d863 Workaround for the mlx kv-cache bug. (#108) Laurent Mazare 2025-08-04 16:37:00 +02:00
9de66be9ff Workaround for the mlx kv-cache bug. mlx-kv-cache-bug Laurent 2025-08-04 16:33:43 +02:00
09468c239a Print the duration of the audio generated so far. (#107) Laurent Mazare 2025-08-04 09:24:31 +02:00
5a719b69c0 Print the duration of the audio generated so far. print-gen laurent 2025-08-04 09:22:36 +02:00
07729ed47e Use the proper repos when vad is on. (#103) Laurent Mazare 2025-08-01 15:55:49 +02:00
35f133508f Use the proper repos when vad is on. vad-repos laurent 2025-07-31 17:47:46 +02:00
af2283de3f Use a streaming input in the rust example. (#102) Laurent Mazare 2025-07-31 17:41:57 +02:00
a1af3a78cb Another formatting tweak. streaming-inptu laurent 2025-07-31 17:40:35 +02:00
ecc002f7af Formatting. laurent 2025-07-31 17:38:34 +02:00
94e271b69a Use a streaming input in the rust example. laurent 2025-07-31 17:37:43 +02:00
7dc926d50c Allow for using local voices in the pytorch examples. (#100) Laurent Mazare 2025-07-31 12:48:05 +02:00
a4e9676e46 Allow for using local voices in the pytorch examples. local-voices laurent 2025-07-31 12:42:35 +02:00
ab8e8c59b7 Bump the version numbers. (#91) Laurent Mazare 2025-07-19 15:57:53 +02:00
31bcee0ed4 Bump the version numbers. bump-versions Laurent 2025-07-19 15:56:12 +02:00
5f17114618 More faq. laurent 2025-07-18 08:43:10 +02:00
405a82ba3f FAQ tweaks. laurent 2025-07-18 08:37:51 +02:00
3b584b100c Sketch a FAQ and add some issue templates. (#88) Laurent Mazare 2025-07-18 08:31:31 +02:00
cd3cc1ce0e Sketch a FAQ and add some issue templates. faq laurent 2025-07-18 08:27:30 +02:00
a98eb94ade Add a streaming example for the mlx tts. (#85) Laurent Mazare 2025-07-16 22:35:44 +02:00
00daf19074 Yet another CI fix. tts-mlx-streaming Laurent 2025-07-16 22:34:42 +02:00
45f01cc617 Formatting fix. Laurent 2025-07-16 22:30:03 +02:00
00b856a94d Fix the CI. Laurent 2025-07-16 22:28:18 +02:00
64c7d55cdb Add a streaming example for the mlx tts. Laurent 2025-07-16 22:26:22 +02:00
a2f031deb5 Fix the pytorch tts streaming example. (#84) Laurent Mazare 2025-07-16 21:07:02 +02:00
87b3c602c9 Edit the readme too. fix-tts-stream laurent 2025-07-16 21:06:05 +02:00
cecbe46d4b Fix the pytorch tts streaming example. laurent 2025-07-16 21:01:16 +02:00
66a33c989f Add a TTS streaming example. (#83) Laurent Mazare 2025-07-16 20:54:13 +02:00
edce117900 Get the streaming example to work. tts-pth-streaming laurent 2025-07-16 20:52:32 +02:00
49ce18be2e Add a TTS streaming example. laurent 2025-07-16 17:23:19 +02:00
89a2ced839 Bump the moshi version. laurent 2025-07-16 16:02:28 +02:00
baf0c75bba VAD support in the mlx-stt example that uses the microphone. Laurent 2025-07-08 16:08:32 +02:00
952319de90 Add a MLX STT example that uses VAD. (#70) Laurent Mazare 2025-07-08 16:04:50 +02:00
5e693debc7 Use the latest moshi-mlx. mlx-vad-example Laurent 2025-07-08 16:02:59 +02:00
e76b2a53ff More MLX VAD example. Laurent 2025-07-08 15:59:42 +02:00
dc6f552a47 VAD support. Laurent 2025-07-08 15:43:38 +02:00
846042d0a4 Merge remote-tracking branch 'origin/main' into mlx-vad-example Laurent 2025-07-08 15:39:38 +02:00
6d3bb6b1f1 Avoid the config override for the extra-heads. laurent 2025-07-08 15:39:17 +02:00
7bf87eaa98 Add a MLX STT example that uses VAD. Laurent 2025-07-08 15:23:20 +02:00
221f3cbad8 Avoid the config override for the extra-heads. vad-python-example laurent 2025-07-08 12:18:28 +02:00
12dbe36b0b Add some VAD to the pytorch speech-to-text example. (#68) Laurent Mazare 2025-07-08 11:30:34 +02:00
c9c4e0a573 Add some VAD to the pytorch speech-to-text example. laurent 2025-07-08 11:28:54 +02:00
cafac63222 Run pre-commit correctly in CI (#66) Václav Volhejn 2025-07-08 10:11:52 +02:00
e497a53f0b Remove intentional error vv/fix-pre-commit Václav Volhejn 2025-07-07 18:05:05 +02:00
2683557584 fix and break Václav Volhejn 2025-07-07 18:03:11 +02:00
7336d7a3da Fix instructions on how to install the Rust server (#65) Václav Volhejn 2025-07-07 18:00:23 +02:00
55632edfb2 Plug Unmute vv/rust-server-instructions Václav Volhejn 2025-07-07 16:50:33 +02:00
c565c7ab6d Fix instructions on Rust server installation Václav Volhejn 2025-07-07 16:49:10 +02:00
70500c620e Add a device argument to the tts pytorch script. (#62) Laurent Mazare 2025-07-07 08:36:47 +02:00
8927c0d9eb Add a device argument to the tts pytorch script. tts-pytorch.device laurent 2025-07-07 08:31:03 +02:00
f8e97aa4f3 fix minor issues with readme commands (#55) Chenghao Mou 2025-07-07 07:18:05 +01:00
91a4d120cb Use moshi 0.2.8. laurent 2025-07-07 08:12:16 +02:00
bfc200f6ee Use bfloat16 rather than half by default. Laurent 2025-07-05 23:02:58 +02:00
f9739881e6 Typo. laurent 2025-07-03 19:02:42 +02:00
99599fa408 Update README.md Alexandre Défossez 2025-07-03 16:15:01 +02:00
3a4165a84f Fix stt_from_file_pytorch (#39) Pierre-Hugues HUSSON 2025-07-03 15:26:34 +02:00
e9bac066ea Update README.md Alexandre Défossez 2025-07-03 15:09:41 +02:00
eae5e17975 Some updates to the colab and script (#38) Alexandre Défossez 2025-07-03 15:06:37 +02:00
5afa2fe656 plop changing_streaming Alexandre Défossez 2025-07-03 15:06:28 +02:00
774ef275a4 plop Alexandre Défossez 2025-07-03 15:04:21 +02:00
080f349403 plop Alexandre Défossez 2025-07-03 15:04:01 +02:00
0b5a5fbed2 plop Alexandre Défossez 2025-07-03 14:55:24 +02:00
31f8746881 some changes Alexandre Défossez 2025-07-03 14:52:02 +02:00
b2416b19dd changing streaming to be robust to repeated generation Alexandre Défossez 2025-07-03 14:38:25 +02:00
c1d248abba Fix text tokenizer path (#36) Václav Volhejn 2025-07-03 14:27:06 +02:00
c6f262346f Don't install moshi from Git (#37) Václav Volhejn 2025-07-03 13:37:38 +02:00
d186eb9594 Remove commented-out invalid message send in websocket_client vv/moshi-version Václav Volhejn 2025-07-03 13:32:56 +02:00
8e07dd55a3 Don't install moshi from Git Václav Volhejn 2025-07-03 13:32:27 +02:00
3573ee90af Oops. laurent 2025-07-03 13:08:00 +02:00
25574aa104 Fixes for the notebook. laurent 2025-07-03 13:05:00 +02:00
1cd9529f65 Json fix. laurent 2025-07-03 12:57:22 +02:00
0ee2354176 Chunk decoding in the pth notebook. laurent 2025-07-03 12:56:00 +02:00
dc8bffabe0 Remove the dataset bit. laurent 2025-07-03 12:48:04 +02:00
5f8e924176 Streaming output for the pytorch tts example. (#33) Laurent Mazare 2025-07-03 11:05:06 +02:00
c970b86088 Run the pre-commit hooks. pth-tts-streaming laurent 2025-07-03 11:04:24 +02:00
cbd2792d29 Fix text tokenizer path vv/tokenizer-repo Václav Volhejn 2025-07-03 10:04:55 +02:00
d3bed09f9a Pin the moshi_mlx version. (#35) Laurent Mazare 2025-07-03 09:53:53 +02:00
e74fc09e37 Pin the moshi_mlx version. mlx-0.2.9 Laurent 2025-07-03 09:51:48 +02:00
9ba4e88553 Merge stuff. laurent 2025-07-03 09:49:24 +02:00