Files
blind/audio_2/tts/v4_voice_samples/README.md
T
xsl 12eb1eab44 VoxCPM2 v5 音色克隆:脚本、参考音频、游戏对白与状态文档
通过 Voice Design 试听选定各角色参考音色,Style Control 全量克隆并部署到 audio/mp3/tts;清理未引用音频与 audio_1 旧产物。
2026-05-24 00:34:50 +08:00

52 lines
1.2 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
# Voice Design 试听样本(v4
每角色 **5 条**约 **7–10 秒**试听,用不同提示词生成不同音色。
选定后注册为 `voice/{角色}/` 参考音频,再跑全量 Style Control 克隆。
## 试听目录
```
audio_2/tts/v4_voice_samples/
linxia/ linxia_v01.mp3 … v05.mp3
mom/
azhe/
xiaomei/
xiaomei_drunk/
zhounan/
hr/
anan/
dorm_a/ dorm_b/ dorm_c/
delivery/ ← 5 条均为男声,语速已调慢
self_3y/
```
完整清单(含时长与提示词):`samples_manifest.tsv`
## 外卖小哥(delivery)变体说明
| 编号 | 侧重点 |
|------|--------|
| v01 | 北方男声,中低音,**语速适中偏慢**,礼貌直接 |
| v02 | 偏低沉稳,**语速慢**,熟手外卖员 |
| v03 | 普通青年男声,语速正常 |
| v04 | 略粗中音,语速适中,不催促 |
| v05 | 清爽青年音,**语速偏慢**,清楚有条理 |
## 选定后告诉我
格式示例:
```
linxia v02, mom v01, azhe v03, delivery v02, ...
```
我会把对应 `.wav` 复制到 `voice/{角色}/`,再全量生成最终对白。
## 重新生成某角色样本
```bash
VOXCPM_PORT=8002 python audio_2/scripts/03_gen_voice_samples.py --role delivery --force
```
提示词配置:`audio_2/voice_sample_prompts.py`