 Kedara Studios
|
7bf2c43a2e
|
feat: add two-phase streaming for reduced first-chunk latency
- Phase 1: aggressive emit/decode settings for fast first chunk
- Phase 2: stable settings with optimized decode for quality
- New parameters: first_chunk_emit_every, first_chunk_decode_window, first_chunk_frames
- Phase 1 disables torch.compile to allow flexible window sizes
|
2026-02-02 20:27:15 +01:00 |
|
|
dffdeeq
|
01b51f0cb1
|
README update
|
2026-01-28 02:26:50 +04:00 |
|
|
Andrew
|
8c533a60e0
|
Merge pull request #3 from dffdeeq/fix/streaming-startup-artifacts
fix(streaming): add ref_code context for stable first chunks
|
2026-01-27 19:49:39 +04:00 |
|
|
dffdeeq
|
b7f12c4941
|
fix(streaming): add ref_code context for stable first chunks
|
2026-01-27 19:14:52 +04:00 |
|
|
dffdeeq
|
990bfa2766
|
Revert upstream changes that broke inference speed
|
2026-01-27 11:18:11 +04:00 |
|
|
dffdeeq
|
e5662324c6
|
minor
|
2026-01-26 23:24:18 +04:00 |
|
|
Andrew
|
66f26eb640
|
Merge pull request #2 from dffdeeq/feature/inference-speed-up
reference changed
|
2026-01-26 23:17:05 +04:00 |
|
|
Andrew
|
ee3bf7a862
|
Merge branch 'main' into feature/inference-speed-up
|
2026-01-26 23:16:53 +04:00 |
|
|
dffdeeq
|
b5a75200d9
|
reference changed
|
2026-01-26 23:14:50 +04:00 |
|
|
Andrew
|
648b7100bd
|
benchmarks update
Fixed formatting issues and improved clarity in the README.
|
2026-01-26 17:54:30 +04:00 |
|
|
Andrew
|
45d455dd3f
|
Merge pull request #1 from dffdeeq/feature/inference-speed-up
Feature/inference speed up
|
2026-01-26 16:57:34 +04:00 |
|
|
dffdeeq
|
f2ae2b039c
|
no streaming optimization
|
2026-01-26 15:52:39 +04:00 |
|
|
dffdeeq
|
d8488d2727
|
,,,
|
2026-01-26 06:16:51 +04:00 |
|
|
Andrew
|
e03d44f3f4
|
README.md update
|
2026-01-25 23:45:27 +04:00 |
|
|
Andrew
|
d7fd4bb1ae
|
Update language and text for voice cloning example
|
2026-01-25 23:33:58 +04:00 |
|
|
Andrew
|
ec708743ea
|
Merge branch 'QwenLM:main' into main
|
2026-01-25 23:26:36 +04:00 |
|
|
Xiong Wang
|
1ab0dd7535
|
update: update readme
|
2026-01-26 00:11:02 +08:00 |
|
|
Xiong Wang
|
3b30a4e509
|
update: update readme
|
2026-01-24 21:44:35 +08:00 |
|
|
Xiong Wang
|
c25ce958ea
|
Merge pull request #15 from vasqu/fix-fa-flags
[`FA`] Fixup wrong flags and loading logic
|
2026-01-24 16:32:32 +08:00 |
|
|
vasqu
|
ab0f77849b
|
fix loading logic
|
2026-01-23 10:48:11 +01:00 |
|
|
vasqu
|
587ebf6389
|
fixup wrong flags
|
2026-01-23 10:18:08 +01:00 |
|
|
Xiong Wang
|
0c6a7cbb6c
|
update: modify default non_streaming_mode
|
2026-01-23 14:08:26 +08:00 |
|
|
Andrew
|
6c07ea2a85
|
Merge branch 'QwenLM:main' into main
|
2026-01-23 09:20:45 +04:00 |
|
|
Xiong Wang
|
8a98526da0
|
fix and update: fix finetuning/prepare_data.py and update citation
|
2026-01-23 13:19:19 +08:00 |
|
|
dffdeeq
|
2331298c38
|
add streaming TTS generation
- stream_generate_pcm() | real-time
- stream_generate_voice_clone()
|
2026-01-23 08:58:38 +04:00 |
|
|
Xiong Wang
|
d5c446171a
|
fix: hf speech_tokenizer download bug.
|
2026-01-22 23:34:21 +08:00 |
|
|
Xiong Wang
|
d8146d5305
|
Initial commit
|
2026-01-22 17:57:04 +08:00 |
|