CoolFace
Modelpublic

MoYoYoTech/VoiceDialogue

sourceHugging Facemitupdated 3mo agoView on Hugging Face
6likes45downloads
50 commits on main
9dca1cb3mo ago

README: drop DMG section (installer not publicly hosted)

hzeng412, Claude Fable 5
040114b3mo ago

README: fix model-download instructions for clone install

hzeng412, Claude Fable 5
2e4bd5d4mo ago

Bundle Qwen3-ASR weights in packaged app; slim PyInstaller assets

hzeng412, Claude Fable 5
21dbad44mo ago

ASR: use auto language detection to reject non-speech (fixes hallucination)

hzeng412, Claude Fable 5
3c704984mo ago

Make Qwen3-ASR the default on main; bump to 1.2.0

hzeng412, Claude Fable 5
a3bf5594mo ago

Add audio output device (speaker) selection

hzeng412, Claude Fable 5
c3f50824mo ago

Show the active ASR engine in the UI

hzeng412, Claude Fable 5
e73f2e44mo ago

Add script to convert TTS .bin weights to safetensors

hzeng412, Claude Fable 5
a17b4204mo ago

Fix websocket route registration for starlette >= 1.0

hzeng412, Claude Fable 5
417ffa14mo ago

ASR: add Qwen3-ASR engine, default zh/en to it on this branch

hzeng412, Claude Fable 5
3c3f6104mo ago

README: lead with GUI (clone-and-run) quick start, move CLI after

hzeng412, Claude Opus 4.8 (1M context)
89dd8964mo ago

Electron: open external links in system browser

hzeng412, Claude Opus 4.8 (1M context)
45d25d14mo ago

Add About tab in settings (models used + repo links)

hzeng412, Claude Opus 4.8 (1M context)
afd6c894mo ago

Add external mic array support, UI i18n, glass settings redesign, and fixes

hzeng412, Claude Opus 4.8 (1M context)
2f3888a1y ago

Update project docs.

liumaolin
60f82381y ago

refactor(core): Architecturally decouple Audio, ASR, and TTS modules

liumaolin
2baeba21y ago

简化LLM模型路径管理。

liumaolin
b319d191y ago

打包Python时默认激活Python虚拟环境

liumaolin
00c731a1y ago

修复electron app打包时,需要notary两次的问题

liumaolin
0d407891y ago

[fix]: notarize debug info.

Xin Zhang
f7c398e1y ago

更新前端资源文件

liumaolin
68831441y ago

[fix]: role bugfix.

Xin Zhang
d2053831y ago

为LLMResponseGenerator类添加TaskStatusMixin混入,以增强任务有效性检查功能,确保在生成回答时任务状态的有效性。

liumaolin
eeb7fe51y ago

Update frontend assets.

liumaolin
4029b291y ago

Merge branches 'main' and 'main' of hf.co:MoYoYoTech/VoiceDialogue

Xin Zhang
797cd521y ago

[fix]: voice selection bugfix.

Xin Zhang
2c7e7421y ago

更新音频任务处理逻辑,使用模型副本替代原始任务对象,以确保数据一致性和完整性。

liumaolin
e3d17e21y ago

重构处理LLM生成chunk内容的方法,以提高文本生成的准确性。

liumaolin
08fc60d1y ago

去除对 LLM 输出的处理

liumaolin
c531bfc1y ago

更新LLM模型参数,调整top_p值为0.8,并添加max_tokens参数,设置为32768。

liumaolin
bbf79f21y ago

更新用户配置文件,修复logger导入路径以确保正确加载日志记录功能。

liumaolin
fb2e3d61y ago

更新加载页面图标路径,添加新的图标文件至assets/images目录。

liumaolin
1cc5d821y ago

更新前端文件

liumaolin
31e11af1y ago

[fix]: improve style.

Xin Zhang
8ba9c841y ago

[fix]: remove frontend ref audio.

Xin Zhang
f71be4e1y ago

Merge branch 'main' of hf.co:MoYoYoTech/VoiceDialogue

Xin Zhang
d35a0ba1y ago

[feature]: update prompt config.

Xin Zhang
226effc1y ago

添加获取TTS模型参考音频的API路由,支持通过模型ID获取音频文件。

liumaolin
8ead3061y ago

Merge branch 'main' of hf.co:MoYoYoTech/VoiceDialogue

Xin Zhang
73e6ae01y ago

[feature]:prompt config.

Xin Zhang
52585381y ago

更新API文档

liumaolin
d12af4f1y ago

缓存user_prompts以提高性能

liumaolin
22a06c41y ago

调整prompt的逻辑,确保展示的提示词中不包含/no_think指令

liumaolin
15ab4781y ago

添加设置相关的API路由

liumaolin
18161301y ago

在测试文件中添加对中文和英文系统提示的导入检查,以确保在未定义时能够正确加载配置。

liumaolin
310c0ba1y ago

调整LLM prompt的存放位置

liumaolin
b7ae8391y ago

增强任务有效性检查,添加日志记录以便于调试。

liumaolin
3a6b6fc1y ago

在生成TTS音频时始终记录生成信息。

liumaolin
259c23b1y ago

在`SpeechStateMonitor`类的初始化过程中调用该方法,确保在创建新语音任务时中断任务ID被正确重置。

liumaolin
5f0761d1y ago

更新electron-app的package.json文件,设置版本号为日期。

liumaolin