Issue Station
Sign In
Tools hereToolbox

openclaw - 💡(How to fix) Fix [Bug] Telegram /tts audio reaches TTS path but Edge provider times out; no voice note delivered on v2026.3.13 [1 comments, 2 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
openclaw/openclaw#51466•Fetched 2026-04-08 01:10:51
View on GitHub
Comments
1
Participants
2
Timeline
1
Reactions
0
Author
spadesvspadesv
Participants
RyceRycespadesvspadesv
Timeline (top)
commented ×1

On OpenClaw v2026.3.13, Telegram TTS does not successfully return a voice note in my environment.

The key point is:

  • TTS is not completely unsupported on Telegram
  • /tts audio clearly enters the TTS path
  • but the free Edge/Microsoft provider times out, so no audio is produced and no Telegram voice note is delivered

Error Message

Telegram returned this error message: ❌ Error generating audio: TTS conversion failed: edge: Timed out; openai: no API key; elevenlabs: no API key

  • Telegram only shows text/error output ❌ Error generating audio: TTS conversion failed: edge: Timed out; openai: no API key; elevenlabs: no API key

Root Cause

On OpenClaw v2026.3.13, Telegram TTS does not successfully return a voice note in my environment.

The key point is:

  • TTS is not completely unsupported on Telegram
  • /tts audio clearly enters the TTS path
  • but the free Edge/Microsoft provider times out, so no audio is produced and no Telegram voice note is delivered

Code Example

{
  "messages": {
    "tts": {
      "auto": "always",
      "provider": "edge",
      "edge": {
        "enabled": true,
        "voice": "zh-CN-XiaochenNeural",
        "lang": "zh-CN"
      }
    }
  }
}

---

/tts audio 你好，这是一次 TTS 强制测试
RAW_BUFFERClick to expand / collapse

Summary

On OpenClaw v2026.3.13, Telegram TTS does not successfully return a voice note in my environment.

The key point is:

  • TTS is not completely unsupported on Telegram
  • /tts audio clearly enters the TTS path
  • but the free Edge/Microsoft provider times out, so no audio is produced and no Telegram voice note is delivered

Environment

  • OpenClaw version: v2026.3.13
  • Channel: Telegram
  • OS: Linux
  • Config file: /root/.openclaw/openclaw.json

What I tested

I first tested auto TTS with:

{
  "messages": {
    "tts": {
      "auto": "always",
      "provider": "edge",
      "edge": {
        "enabled": true,
        "voice": "zh-CN-XiaochenNeural",
        "lang": "zh-CN"
      }
    }
  }
}

I also tested auto: "inbound" before that.

In both cases, Telegram replies remained text-only in practice.

Then I ran an explicit command in Telegram:

/tts audio 你好，这是一次 TTS 强制测试

Telegram returned this error message:

❌ Error generating audio: TTS conversion failed: edge: Timed out; openai: no API key; elevenlabs: no API key

Expected behavior

  • /tts audio ... should generate speech and send a Telegram voice note
  • messages.tts.auto = "always" or "inbound" should trigger outbound audio on Telegram
  • Telegram should receive the native round voice-note bubble

Actual behavior

  • /tts audio ... does not deliver any voice note
  • Telegram only shows text/error output
  • The Edge provider appears to time out before audio is returned

Important observation

Please note that this is not the same as "Telegram does not support TTS at all".

The explicit /tts audio command produces:

❌ Error generating audio: TTS conversion failed: edge: Timed out; openai: no API key; elevenlabs: no API key

So the request appears to reach the TTS layer, but the Edge provider fails before Telegram can send audio.

Additional notes

  • No local TTS prefs file existed on this node: ~/.openclaw/settings/tts.json was absent, so there was no prefs override of messages.tts
  • After testing, I restored the config to default by removing messages.tts

Question

Is this a known regression/timeout issue with the Edge/Microsoft TTS path on Telegram in v2026.3.13?

Also, is there any recommended timeout/output-format adjustment for Telegram + Edge TTS in this version?

extent analysis

Fix Plan

To resolve the issue with the Edge/Microsoft TTS provider timing out on Telegram in OpenClaw v2026.3.13, follow these steps:

  • Increase the timeout value: Modify the openclaw.json configuration file to increase the timeout for the Edge TTS provider. Add a timeout field under the edge section:
{
  "messages": {
    "tts": {
      "auto": "always",
      "provider": "edge",
      "edge": {
        "enabled": true,
        "voice": "zh-CN-XiaochenNeural",
        "lang": "zh-CN",
        "timeout": 30000 // 30 seconds
      }
    }
  }
}
  • Specify an output format: Ensure that the output format is set to a compatible format for Telegram voice notes. Add an outputFormat field under the edge section:
{
  "messages": {
    "tts": {
      "auto": "always",
      "provider": "edge",
      "edge": {
        "enabled": true,
        "voice": "zh-CN-XiaochenNeural",
        "lang": "zh-CN",
        "timeout": 30000,
        "outputFormat": "ogg"
      }
    }
  }
}
  • Verify API keys for alternative providers: If you have API keys for other T

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

FAQ

Expected behavior

  • /tts audio ... should generate speech and send a Telegram voice note
  • messages.tts.auto = "always" or "inbound" should trigger outbound audio on Telegram
  • Telegram should receive the native round voice-note bubble
#api#ssr#installation#tensor shape#autograd error#latency issue#model loading#dependency error#configuration error

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING