dsh-audiogen 0.4.22 → 0.4.24
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +2 -2
- package/README.zh-CN.md +2 -2
- package/lib/client.js +497 -414
- package/lib/client.js.map +1 -1
- package/lib/index.js +94 -14
- package/package.json +22 -20
- package/skills/sfx/SKILL.md +7 -3
- package/src/agent-audio-tools.ts +3 -3
- package/src/audio-engine.ts +74 -0
- package/src/client/SettingsCard.tsx +2 -1
- package/src/client/audio-toolview.tsx +5 -14
- package/src/client/channels-form.ts +1 -1
- package/src/client/field-specs.ts +56 -4
- package/src/client/index.ts +2 -20
- package/src/client/settings-form.ts +2 -1
- package/src/client/settings-scope.ts +10 -9
- package/src/client/studio-view.tsx +21 -10
- package/src/index.ts +11 -10
- package/src/protocol.ts +4 -4
- package/src/routes.ts +2 -2
package/README.md
CHANGED
|
@@ -15,7 +15,7 @@
|
|
|
15
15
|
|
|
16
16
|
- **Four generation modes**: text-to-speech, music, sound effects, and voice design
|
|
17
17
|
- **Multi-vendor channels in one place**: OpenAI-compatible TTS, MiniMax, ElevenLabs, Stability AI, or any custom OpenAI-compatible / generic POST endpoint
|
|
18
|
-
- **Per-channel model & voice catalogs** with one-click discovery, display aliases, capability categories, and per-model advanced fields (duration, seed, steps, cfg_scale, loop, prompt influence, …)
|
|
18
|
+
- **Per-channel model & voice catalogs** with one-click discovery, display aliases, capability categories, and per-model advanced fields (duration, seed, steps, cfg_scale, loop, prompt influence, and ElevenLabs SFX output format split into format/sample rate/bitrate, combined host-side into the single `output_format`, …)
|
|
19
19
|
- **Model comparison**: run the same prompt across 2–4 models at once with per-model parameter overrides — results are grouped side by side
|
|
20
20
|
- **Prompt enhancement**: rewrite a rough idea into a ready-to-generate description with an LLM (pick any model from *Settings → Models*; falls back to the agent default model)
|
|
21
21
|
- **History with one-click restore**: prompt, config, model set *and the original audio* come back into the panel — no regeneration, no extra cost
|
|
@@ -72,7 +72,7 @@ Restart `dsh web` after install — the sidebar will show the **AI Audio** entry
|
|
|
72
72
|
| --- | --- | --- | --- | --- |
|
|
73
73
|
| TTS | ✅ (8 voices) | ✅ (voices + streams) | — | ✅ |
|
|
74
74
|
| Music | ✅ (`music-3.0` / `music-2.6` / `music-cover`) | ✅ (`music_v2`) | ✅ (`stable-audio-*`) | ✅ (generic POST) |
|
|
75
|
-
| Sound effects | — | ✅ (`eleven_text_to_sound_v2`, loop / prompt influence) | ✅ (`stable-audio-*` — same text-to-audio protocol; auto-detected in both Music and SFX) | ✅ (generic POST) |
|
|
75
|
+
| Sound effects | — | ✅ (`eleven_text_to_sound_v2`, loop / prompt influence / output format as codec+sample rate+bitrate → `output_format`) | ✅ (`stable-audio-*` — same text-to-audio protocol; auto-detected in both Music and SFX) | ✅ (generic POST) |
|
|
76
76
|
| Voice design | ✅ (`/v1/voice_design`) | ✅ (`/v1/text-to-voice/design`) | — | — |
|
|
77
77
|
|
|
78
78
|
## 🤖 Agent usage
|
package/README.zh-CN.md
CHANGED
|
@@ -15,7 +15,7 @@
|
|
|
15
15
|
|
|
16
16
|
- **四种生成模式**:文本转语音(TTS)· 音乐生成 · 音效生成 · 音色设计
|
|
17
17
|
- **多厂商渠道统一管理**:OpenAI 兼容 TTS、MiniMax、ElevenLabs、Stability AI,以及任意自定义 OpenAI 兼容 / 通用 POST 接口
|
|
18
|
-
- **每渠道模型/音色目录**:一键「获取可用模型」、显示名称(alias)、能力分类(语音/音乐/音效/音色设计),按模型展示完整参数(时长、seed、steps、cfg_scale、loop
|
|
18
|
+
- **每渠道模型/音色目录**:一键「获取可用模型」、显示名称(alias)、能力分类(语音/音乐/音效/音色设计),按模型展示完整参数(时长、seed、steps、cfg_scale、loop、提示词影响度,ElevenLabs 音效输出格式拆为「格式/采样率/码率」三个参数、由引擎组合为 `output_format`……)
|
|
19
19
|
- **模型对比**:同一提示词并发生成 2–4 个模型,支持每模型参数覆盖,结果按模型分组并列对比
|
|
20
20
|
- **✨ 提示词增强**:把一句粗略想法扩写成适合生成模型的完整描述,LLM 模型可在「设置 → 模型」中任选(缺省跟随 Agent 默认模型,无需额外 API Key)
|
|
21
21
|
- **历史记录一键恢复**:prompt、配置、模型组合**以及当时的音频**全部回到面板——可直接试听/下载,无需重新生成、不额外消耗
|
|
@@ -72,7 +72,7 @@ dsh plugin --profile web add /path/to/dsh-audiogen
|
|
|
72
72
|
| --- | --- | --- | --- | --- |
|
|
73
73
|
| TTS | ✅(8 种音色) | ✅(音色 + 流式) | — | ✅ |
|
|
74
74
|
| 音乐 | ✅(`music-3.0` / `music-2.6` / `music-cover`) | ✅(`music_v2`) | ✅(`stable-audio-*`) | ✅(通用 POST) |
|
|
75
|
-
| 音效 | — | ✅(`eleven_text_to_sound_v2`,loop / prompt_influence
|
|
75
|
+
| 音效 | — | ✅(`eleven_text_to_sound_v2`,loop / prompt_influence / 输出格式 格式+采样率+码率 → `output_format`) | ✅(`stable-audio-*`,同一 text-to-audio 协议,自动识别为音乐+音效双模式) | ✅(通用 POST) |
|
|
76
76
|
| 音色设计 | ✅(`/v1/voice_design`) | ✅(`/v1/text-to-voice/design`) | — | — |
|
|
77
77
|
|
|
78
78
|
## 🤖 Agent 使用
|