快速单说话人语音合成Fast and High-Quality Zero-Shot Text-to-Speech with Flow Matching
active 2025-06-20 → 2026-07-04 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 380 days from 2025-06-20 to 2026-07-04. Pushes: 23 total, peak 3 in a day. Pull requests: 49 total, peak 7 in a day. Issues: 119 total, peak 8 in a day. Comments: 268 total, peak 16 in a day. Stars: 433 total, peak 35 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| zhu-han | 180 | 12 | 25 | 121 |
| pkufool | 36 | 9 | 12 | 7 |
| Alekksander66 | 20 | 0 | 0 | 17 |
| wincing2 | 18 | 0 | 0 | 7 |
| yaguanghu | 14 | 0 | 0 | 12 |
| blackbird-fish | 9 | 0 | 0 | 7 |
| alpcansoydas | 8 | 0 | 0 | 2 |
| kmmang | 8 | 0 | 0 | 3 |
| eaysu | 8 | 0 | 0 | 1 |
| Respaired | 7 | 0 | 0 | 7 |
| einsqing | 6 | 0 | 0 | 5 |
| Ashish-Patnaik | 5 | 0 | 0 | 4 |
| huanxian | 5 | 0 | 0 | 2 |
| didadida-r | 4 | 0 | 0 | 1 |
| Copilot | 4 | 0 | 0 | 4 |
| wei123-stack | 4 | 0 | 0 | 2 |
| fkmaan-commits | 4 | 0 | 0 | 1 |
| segmentationFaults | 3 | 0 | 0 | 3 |
| sajal-madan-ds | 3 | 0 | 0 | 0 |
| yaozengwei | 3 | 0 | 1 | 2 |
Recent activity
Latest issues, pull requests and releases
- Issue#185Imesh72026-04-22 07:18conditional time masking ratio can be reduced while training?
- Issue comment#18199-song2026-04-22 04:36When training on libritts, the speech generated by inference after training is garbled.
- Issue#185Imesh72026-04-18 08:17conditional time masking ratio can be reduced while training?
- Issue#170eaysu2026-03-09 21:32Multi-Language Finetuning
- Issue#178Alekksander662026-03-06 09:36Finetuning after training from scratch
- Issue comment#175mukherjeesougata-eros2026-02-19 06:51Training restarts from epoch 1 on rerun even though previous epoch checkpoints exist
- Issue#176mukherjeesougata-eros2026-02-18 07:45Experimental setting/setup guidance required for training distill models on 3.7k hours custom dataset
- Issue#171alpcansoydas2026-02-11 11:41Special tokens generation in ZipVoice-Dialogue
- Issue#172kobenaxie2026-01-27 06:15What is the role of feat-scale?
- Issue comment#170zhu-han2026-01-19 02:14Multi-Language Finetuning
- Issue comment#169Shenkailai2026-01-16 10:03中文tts韵律问题
- Issue comment#167zhu-han2026-01-07 11:46验证时显存飙升20G
- Issue#166ppeev0012026-01-04 01:46合成文本尾字吞音问题【GPU与CPU版本表现不一致】
- Issue comment#166zhu-han2025-12-30 04:02合成文本尾字吞音问题【GPU与CPU版本表现不一致】
- Issue#159eaysu2025-12-26 09:43Crackling/harsh artifacts for high‑energy emotions (angry/excited) in voice cloning
- Issue comment#159zhu-han2025-12-26 09:42Crackling/harsh artifacts for high‑energy emotions (angry/excited) in voice cloning
- Issue comment#159zhu-han2025-12-26 09:33Crackling/harsh artifacts for high‑energy emotions (angry/excited) in voice cloning
- Issue comment#159zhu-han2025-12-26 04:07Crackling/harsh artifacts for high‑energy emotions (angry/excited) in voice cloning
- Issue comment#166ppeev0012025-12-25 12:52合成文本尾字吞音问题【GPU与CPU版本表现不一致】
- Issue comment#166zhu-han2025-12-25 07:31合成文本尾字吞音问题【GPU与CPU版本表现不一致】
- Issue comment#164yuekaizhang2025-12-25 03:52关于nvidia-triton工具完善
- Issue comment#165usndangol972025-12-25 03:45Zipvoice Model Training
- Issue comment#163zhu-han2025-12-25 03:12zipvoice-dialog也可以用in-house数据训练吗
- Issue#165usndangol972025-12-24 06:42Zipvoice Model Training
- Issue comment#163kmmang2025-12-23 01:17zipvoice-dialog也可以用in-house数据训练吗
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 433 stars here means stars gained during the window, not the repo's star count.