Issue Station
Sign In
Tools hereToolbox

openclaw - 💡(How to fix) Fix [Bug]: TTS tool reports "Generated audio reply" but does not generate or deliver audio when ElevenLabs is configured (Telegram) [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#63403•Fetched 2026-04-09 07:54:10
View on GitHub
Comments
1
Participants
2
Timeline
3
Reactions
0
Author
projectdivisual-droidprojectdivisual-droid
Participants
mjamivmjamivprojectdivisual-droidprojectdivisual-droid
Timeline (top)
labeled ×2commented ×1

The built-in tts tool returns "Generated audio reply" but no audio file is actually generated or delivered to Telegram.

Root Cause

The built-in tts tool returns "Generated audio reply" but no audio file is actually generated or delivered to Telegram.

Fix Action

Workaround

Calling the ElevenLabs API directly via curl and sending the resulting MP3 with message(action=send, asVoice=true) works perfectly. The ElevenLabs API key and config are valid.

RAW_BUFFERClick to expand / collapse

Bug type

Regression (worked before, now fails)

Beta release blocker

No

Summary

Environment

  • OpenClaw version: 2026.4.8
  • Channel: Telegram
  • TTS provider: ElevenLabs (eleven_multilingual_v2)
  • OS: Linux 6.8.0-107-generic (x64)

Description

The built-in tts tool returns "Generated audio reply" but no audio file is actually generated or delivered to Telegram.

Steps to reproduce

  1. Configure messages.tts.provider: "elevenlabs" with valid API key and voice settings
  2. Add elevenlabs to plugins.allow and plugins.entries
  3. Verify plugin loads (openclaw plugins list shows elevenlabs as loaded)
  4. Send a voice message to trigger TTS auto-reply (auto mode: inbound)
  5. Agent calls tts tool → returns "Generated audio reply"
  6. No audio file is created in /tmp/openclaw/tts-*/ and no voice message is delivered to Telegram

Workaround

Calling the ElevenLabs API directly via curl and sending the resulting MP3 with message(action=send, asVoice=true) works perfectly. The ElevenLabs API key and config are valid.

Related

  • The plugins.allow allowlist was previously blocking the ElevenLabs plugin silently (fixed by adding elevenlabs to the list)
  • Fix #62205 (2026.4.7) addressed plugins.allow blocking STT providers but may not cover TTS providers
  • Files generated before the fix (when fallback to OpenAI TTS was used) show 24kHz mono audio, confirming ElevenLabs was never actually called by the built-in tool

Expected behavior

When ElevenLabs is configured, enabled, loaded, and the tts tool is called, it should generate audio via ElevenLabs and deliver it as a voice note on Telegram.

Steps to reproduce

Environment

  • OpenClaw version: 2026.4.8
  • Channel: Telegram
  • TTS provider: ElevenLabs (eleven_multilingual_v2)
  • OS: Linux 6.8.0-107-generic (x64)

Description

The built-in tts tool returns "Generated audio reply" but no audio file is actually generated or delivered to Telegram.

Steps to reproduce

  1. Configure messages.tts.provider: "elevenlabs" with valid API key and voice settings
  2. Add elevenlabs to plugins.allow and plugins.entries
  3. Verify plugin loads (openclaw plugins list shows elevenlabs as loaded)
  4. Send a voice message to trigger TTS auto-reply (auto mode: inbound)
  5. Agent calls tts tool → returns "Generated audio reply"
  6. No audio file is created in /tmp/openclaw/tts-*/ and no voice message is delivered to Telegram

Workaround

Calling the ElevenLabs API directly via curl and sending the resulting MP3 with message(action=send, asVoice=true) works perfectly. The ElevenLabs API key and config are valid.

Related

  • The plugins.allow allowlist was previously blocking the ElevenLabs plugin silently (fixed by adding elevenlabs to the list)
  • Fix #62205 (2026.4.7) addressed plugins.allow blocking STT providers but may not cover TTS providers
  • Files generated before the fix (when fallback to OpenAI TTS was used) show 24kHz mono audio, confirming ElevenLabs was never actually called by the built-in tool

Expected behavior

When ElevenLabs is configured, enabled, loaded, and the tts tool is called, it should generate audio via ElevenLabs and deliver it as a voice note on Telegram.

Expected behavior

Environment

  • OpenClaw version: 2026.4.8
  • Channel: Telegram
  • TTS provider: ElevenLabs (eleven_multilingual_v2)
  • OS: Linux 6.8.0-107-generic (x64)

Description

The built-in tts tool returns "Generated audio reply" but no audio file is actually generated or delivered to Telegram.

Steps to reproduce

  1. Configure messages.tts.provider: "elevenlabs" with valid API key and voice settings
  2. Add elevenlabs to plugins.allow and plugins.entries
  3. Verify plugin loads (openclaw plugins list shows elevenlabs as loaded)
  4. Send a voice message to trigger TTS auto-reply (auto mode: inbound)
  5. Agent calls tts tool → returns "Generated audio reply"
  6. No audio file is created in /tmp/openclaw/tts-*/ and no voice message is delivered to Telegram

Workaround

Calling the ElevenLabs API directly via curl and sending the resulting MP3 with message(action=send, asVoice=true) works perfectly. The ElevenLabs API key and config are valid.

Related

  • The plugins.allow allowlist was previously blocking the ElevenLabs plugin silently (fixed by adding elevenlabs to the list)
  • Fix #62205 (2026.4.7) addressed plugins.allow blocking STT providers but may not cover TTS providers
  • Files generated before the fix (when fallback to OpenAI TTS was used) show 24kHz mono audio, confirming ElevenLabs was never actually called by the built-in tool

Expected behavior

When ElevenLabs is configured, enabled, loaded, and the tts tool is called, it should generate audio via ElevenLabs and deliver it as a voice note on Telegram.

Actual behavior

Environment

  • OpenClaw version: 2026.4.8
  • Channel: Telegram
  • TTS provider: ElevenLabs (eleven_multilingual_v2)
  • OS: Linux 6.8.0-107-generic (x64)

Description

The built-in tts tool returns "Generated audio reply" but no audio file is actually generated or delivered to Telegram.

Steps to reproduce

  1. Configure messages.tts.provider: "elevenlabs" with valid API key and voice settings
  2. Add elevenlabs to plugins.allow and plugins.entries
  3. Verify plugin loads (openclaw plugins list shows elevenlabs as loaded)
  4. Send a voice message to trigger TTS auto-reply (auto mode: inbound)
  5. Agent calls tts tool → returns "Generated audio reply"
  6. No audio file is created in /tmp/openclaw/tts-*/ and no voice message is delivered to Telegram

Workaround

Calling the ElevenLabs API directly via curl and sending the resulting MP3 with message(action=send, asVoice=true) works perfectly. The ElevenLabs API key and config are valid.

Related

  • The plugins.allow allowlist was previously blocking the ElevenLabs plugin silently (fixed by adding elevenlabs to the list)
  • Fix #62205 (2026.4.7) addressed plugins.allow blocking STT providers but may not cover TTS providers
  • Files generated before the fix (when fallback to OpenAI TTS was used) show 24kHz mono audio, confirming ElevenLabs was never actually called by the built-in tool

Expected behavior

When ElevenLabs is configured, enabled, loaded, and the tts tool is called, it should generate audio via ElevenLabs and deliver it as a voice note on Telegram.

OpenClaw version

2026.4.8

Operating system

Linux 6.8.0-107-generic (x64)

Install method

No response

Model

opus4.6

Provider / routing chain

telegram

Additional provider/model setup details

No response

Logs, screenshots, and evidence

Impact and severity

No response

Additional information

No response

extent analysis

TL;DR

The issue can be worked around by directly calling the ElevenLabs API via curl and sending the resulting MP3 with message(action=send, asVoice=true).

Guidance

  • Verify that the ElevenLabs plugin is correctly configured and loaded by checking the plugins.allow and plugins.entries lists.
  • Check the OpenClaw version and ensure that it is up-to-date, as the issue may be related to a bug that was fixed in a later version.
  • Test the ElevenLabs API directly using curl to confirm that it is working correctly and generating audio files.
  • Investigate the plugins.allow allowlist and its impact on TTS providers, as it may be silently blocking the ElevenLabs plugin.

Example

curl -X POST \
  https://api.elevenlabs.io/v1/text-to-speech \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{"text": "Hello, world!", "voice": "your_voice"}' \
  -o output.mp3

Then, send the resulting MP3 with message(action=send, asVoice=true).

Notes

The issue may be related to a bug in the OpenClaw version 2026.4.8, and upgrading to a later version may resolve the issue. Additionally, the plugins.allow allowlist may need to be updated to include TTS providers.

Recommendation

Apply the workaround by directly calling the ElevenLabs API via curl and sending the resulting MP3 with message(action=send, asVoice=true), as this has been confirmed to work perfectly.

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

When ElevenLabs is configured, enabled, loaded, and the tts tool is called, it should generate audio via ElevenLabs and deliver it as a voice note on Telegram.

#api#model save/load#optimization#mixed precision#training loop

Still need to ship something?

×6

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

Back to top recommendations

TRENDING