具备细粒度语音编辑能力的一款AI语音修复模型：PlayDiffusion，修改后的语音听上去可与原语音无缝衔接对于做语音播报、或者有声读物，这种需要频繁修改语音内容的场景来说比较实用 支持精准修改，比如对语音中的单个词进行精准修改，把一句话中的“小明”改成“小红” 其在编辑过程中保留了上下文信息，使得修改后的语音与原语音间能保持平滑衔接 通过保留说话者�
Python · active 2025-06-02 → 2026-04-18 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 321 days from 2025-06-02 to 2026-04-18. Pushes: 14 total, peak 3 in a day. Pull requests: 11 total, peak 3 in a day. Issues: 10 total, peak 3 in a day. Comments: 37 total, peak 7 in a day. Stars: 316 total, peak 90 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 |
|---|---|---|---|---|
| YavorGIvanov | 32 | 6 | 3 | 16 |
| bryananderson | 13 | 5 | 2 | 5 |
| Maki9009 | 3 | 0 | 0 | 3 |
| Ssviki | 3 | 0 | 0 | 2 |
| gianpaj | 3 | 0 | 0 | 1 |
| dependabot[bot] | 3 | 0 | 3 | 0 |
| b83 | 2 | 0 | 0 | 1 |
| sgriffey | 2 | 2 | 0 | 0 |
| rikabi89 | 2 | 0 | 0 | 2 |
| hadifarnoud | 2 | 0 | 0 | 1 |
| hanifalisohag | 2 | 0 | 1 | 1 |
| rasenganai | 2 | 0 | 0 | 2 |
| facebook-github-bot | 2 | 1 | 1 | 0 |
| jon-play | 1 | 0 | 1 | 0 |
| coezbek | 1 | 0 | 0 | 1 |
| dconsorte | 1 | 0 | 0 | 1 |
| metantonio | 1 | 0 | 0 | 0 |
| KonstantineGoudz | 1 | 0 | 0 | 1 |
Recent activity
Latest issues, pull requests and releases
- Pull request#26dependabot[bot]2026-04-18 02:15
- Pull request#26dependabot[bot]2026-04-18 02:15
- Pull request#26dependabot[bot]2026-04-18 02:15
- Issue comment#20coezbek2025-10-17 21:11BpeVoiceTokenizer can only handle absolute paths...
- Pull request#19YavorGIvanov2025-10-01 11:44
- Pull request#17YavorGIvanov2025-08-19 15:40
- Pull request#18facebook-github-bot2025-08-14 20:41
- Issue comment#16YavorGIvanov2025-08-09 15:33Docker install error
- Issue comment#15b832025-08-08 20:09Is it possible to finetune this model for another language?
- Issue comment#15YavorGIvanov2025-08-08 19:02Is it possible to finetune this model for another language?
- Issue comment#15hadifarnoud2025-08-08 14:06Is it possible to finetune this model for another language?
- Issue comment#11YavorGIvanov2025-08-08 10:56Slow Audio Generation – Output Voice Slower Than Input
- Issue#11YavorGIvanov2025-08-08 10:56Slow Audio Generation – Output Voice Slower Than Input
- Issue#7YavorGIvanov2025-08-08 10:54Add local transcription support
- Issue comment#15YavorGIvanov2025-08-08 10:52Is it possible to finetune this model for another language?
- Issue#15YavorGIvanov2025-08-08 10:52Is it possible to finetune this model for another language?
- Issue#15b832025-08-06 22:47Is it possible to finetune this model for another language?
- Issue comment#9dconsorte2025-07-08 15:515090 Support
- Issue comment#13bryananderson2025-06-21 18:56Added local transcription support using whisper-timestamped library
- Pull request#14bryananderson2025-06-21 18:55
- Issue comment#3Maki90092025-06-21 18:37Add Windows support
- Issue comment#3Maki90092025-06-21 15:31Add Windows support
- Issue comment#3Maki90092025-06-17 19:12Add Windows support
- Issue comment#13hanifalisohag2025-06-12 06:36Added local transcription support using whisper-timestamped library
- Issue comment#13bryananderson2025-06-12 01:18Added local transcription support using whisper-timestamped library
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 316 stars here means stars gained during the window, not the repo's star count.