# Send Telegram Multilingual Voice Reply to your agent
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
## Fast path
- Download the package from Yavira.
- Extract it into a folder your agent can access.
- Paste one of the prompts below and point your agent at the extracted folder.
## Suggested prompts
### New install

```text
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete.
```
### Upgrade existing

```text
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run.
```
## Machine-readable fields
```json
{
  "schemaVersion": "1.0",
  "item": {
    "slug": "telegram-multilingual-voice-reply",
    "name": "Telegram Multilingual Voice Reply",
    "source": "tencent",
    "type": "skill",
    "category": "AI 智能",
    "sourceUrl": "https://clawhub.ai/pengling9405/telegram-multilingual-voice-reply",
    "canonicalUrl": "https://clawhub.ai/pengling9405/telegram-multilingual-voice-reply",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadUrl": "/downloads/telegram-multilingual-voice-reply",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=telegram-multilingual-voice-reply",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "packageFormat": "ZIP package",
    "primaryDoc": "SKILL.md",
    "includedAssets": [
      "SKILL.md",
      "references/qwen3-asr-notes.md",
      "scripts/mlx_asr.py",
      "scripts/mlx_tts_voice.py"
    ],
    "downloadMode": "redirect",
    "sourceHealth": {
      "source": "tencent",
      "slug": "telegram-multilingual-voice-reply",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-05-04T01:26:37.573Z",
      "expiresAt": "2026-05-11T01:26:37.573Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=telegram-multilingual-voice-reply",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=telegram-multilingual-voice-reply",
        "contentDisposition": "attachment; filename=\"telegram-multilingual-voice-reply-0.1.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null,
        "slug": "telegram-multilingual-voice-reply"
      },
      "scope": "item",
      "summary": "Item download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this item.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/telegram-multilingual-voice-reply"
    },
    "validation": {
      "installChecklist": [
        "Use the Yavira download entry.",
        "Review SKILL.md after the package is downloaded.",
        "Confirm the extracted package contains the expected setup assets."
      ],
      "postInstallChecks": [
        "Confirm the extracted package includes the expected docs or setup files.",
        "Validate the skill or prompts are available in your target agent workspace.",
        "Capture any manual follow-up steps the agent could not complete."
      ]
    }
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/telegram-multilingual-voice-reply",
    "downloadUrl": "https://openagent3.xyz/downloads/telegram-multilingual-voice-reply",
    "agentUrl": "https://openagent3.xyz/skills/telegram-multilingual-voice-reply/agent",
    "manifestUrl": "https://openagent3.xyz/skills/telegram-multilingual-voice-reply/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/telegram-multilingual-voice-reply/agent.md"
  }
}
```
## Documentation

### Telegram Multilingual Voice Reply（多语言语音智能回复）

目标：在 Telegram 跟 OpenClaw 聊天时实现“语音优先 + 多语言跟随”的智能回复：

你发文字 → 默认只发文字回复（不触发 ASR；默认也不触发 TTS）。
你发语音 → 本地 STT 转写后理解语义，生成“有意义的回复”。默认用 一条 Voice Message + caption 发送（语音+文字同条，且内容一致）。
语言跟随：你用中文我用中文；你用英文我用英文；除非你明确要求指定语言（也支持方言/口吻要求，但 TTS 受模型限制）。

### Default models

STT 默认：mlx-community/Qwen3-ASR-0.6B-8bit
（可选）Forced alignment：mlx-community/Qwen3-ForcedAligner-0.6B-8bit

需要模型清单/更多上游用法：references/qwen3-asr-notes.md

### A) 输入是文字（text message）

永不触发 ASR。
默认只发文字回复。
只有当用户明确说“用语音回复/发语音/voice reply”时：

生成回复文本 reply_text
生成 voice note（见下方“Telegram 单条消息”）
发送 一条 Voice Message（caption=reply_text，且与语音内容一致）

### B) 输入是语音（voice note / audio）

默认必走 ASR（否则无法“智能回复”）。
然后根据用户意图决定是否触发 TTS：

用户说“只要文字/文字回复/不要语音” → ASR + 文字回复（不 TTS）。
否则（默认真实场景） → ASR + 生成回复 + 单条 voice+caption（走 TTS）。

语音→转写（本地、Apple Silicon）

python3 scripts/mlx_asr.py --audio /path/to/audio.ogg --language Chinese

注：Telegram 常见是 ogg/opus。mlx_asr.py 已支持 CLI fallback，并会在需要时用 ffmpeg 转 wav。

Telegram 单条消息（语音 + 文字在同一条）

Telegram 支持给 voice note 附带 caption，使得“语音条 + 下方文字”显示为同一条消息（你截图里的效果）。

用 scripts/mlx_tts_voice.py 生成 Telegram 友好的 .ogg/opus：

python3 scripts/mlx_tts_voice.py --text "<reply_text>" --out /tmp/reply.ogg

用 OpenClaw message 工具发送（voice note + caption）：

asVoice: true
path/filePath: 上一步生成的 reply.ogg
caption: 同一段 reply_text（必须与语音内容一致）

说明：OpenClaw 的 tts 工具通常会把语音作为单独消息发送，无法保证 caption 同条承载。

### Options / overrides

用户说“只用文字回复/不要语音” → 只发文字。
用户说“只用语音回复/不要文字” → 只发语音（仍然保持内容一致）。
用户说“用英文/用中文回复” → 覆盖默认语言跟随策略。

### Troubleshooting

failed to import mlx_audio：当前 python3 环境里没有安装到 mlx_audio。请用安装 mlx_audio 的同一个解释器/venv 运行。
音频格式问题：Telegram 常见 voice note 是 ogg/opus。如转写失败，优先把音频转成 wav 再试（或确保系统具备相应解码能力）。
## Trust
- Source: tencent
- Verification: Indexed source record
- Publisher: pengling9405
- Version: 0.1.0
## Source health
- Status: healthy
- Item download looks usable.
- Yavira can redirect you to the upstream package for this item.
- Health scope: item
- Reason: direct_download_ok
- Checked at: 2026-05-04T01:26:37.573Z
- Expires at: 2026-05-11T01:26:37.573Z
- Recommended action: Download for OpenClaw
## Links
- [Detail page](https://openagent3.xyz/skills/telegram-multilingual-voice-reply)
- [Send to Agent page](https://openagent3.xyz/skills/telegram-multilingual-voice-reply/agent)
- [JSON manifest](https://openagent3.xyz/skills/telegram-multilingual-voice-reply/agent.json)
- [Markdown brief](https://openagent3.xyz/skills/telegram-multilingual-voice-reply/agent.md)
- [Download page](https://openagent3.xyz/downloads/telegram-multilingual-voice-reply)