Issue Station
Sign In
Tools hereToolbox

openclaw - ๐Ÿ’ก(How to fix) Fix Bug: TTS tool reports success but no audio file is generated or delivered on WhatsApp [1 participants]

Official PRs (โ€ฆ)
ON THIS PAGE

Recommended Tools

ร—6

Utilities matched from this issueโ€™s tags and category โ€” 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 ยท Quick feedback

Loadingโ€ฆ
GitHub stats
openclaw/openclaw#59919โ€ขFetched 2026-04-08 02:38:47
View on GitHub
Comments
0
Participants
1
Timeline
0
Reactions
0
Author
fray-aifray-ai
Participants
fray-aifray-ai

The tts tool returns "Generated audio reply." but no audio file is actually produced or delivered to the WhatsApp chat. The user receives nothing.

Error Message

  • No error in logs

Root Cause

The tts tool returns "Generated audio reply." but no audio file is actually produced or delivered to the WhatsApp chat. The user receives nothing.

Fix Action

Workaround

Using sag CLI (ElevenLabs TTS) to generate the file manually, then sending via message(action=send, filePath=..., asVoice=true) works correctly:

sag "text" --lang es -o /path/to/output.opus
# Then send via message tool with asVoice=true

This confirms:

  • ElevenLabs API key and config are valid
  • WhatsApp audio sending works
  • The issue is in the TTS tool's file generation/delivery pipeline

Code Example

{
  "messages": {
    "tts": {
      "auto": "tagged",
      "mode": "all",
      "provider": "elevenlabs",
      "providers": {
        "elevenlabs": {
          "voiceId": "kulszILr6ees0ArU8miO",
          "modelId": "eleven_flash_v2_5",
          "languageCode": "es",
          "voiceSettings": {
            "stability": 0.5,
            "similarityBoost": 0.75,
            "speed": 1
          }
        }
      }
    }
  }
}

---

sag "text" --lang es -o /path/to/output.opus
# Then send via message tool with asVoice=true
RAW_BUFFERClick to expand / collapse

Description

The tts tool returns "Generated audio reply." but no audio file is actually produced or delivered to the WhatsApp chat. The user receives nothing.

Environment

  • OpenClaw version: 2026.4.2 (also reproduced on 2026.4.1)
  • Channel: WhatsApp (Baileys)
  • OS: Ubuntu Linux 6.17.0-1007-aws (x64)
  • Node: v25.4.0

TTS Configuration

{
  "messages": {
    "tts": {
      "auto": "tagged",
      "mode": "all",
      "provider": "elevenlabs",
      "providers": {
        "elevenlabs": {
          "voiceId": "kulszILr6ees0ArU8miO",
          "modelId": "eleven_flash_v2_5",
          "languageCode": "es",
          "voiceSettings": {
            "stability": 0.5,
            "similarityBoost": 0.75,
            "speed": 1
          }
        }
      }
    }
  }
}

Steps to Reproduce

  1. Configure TTS with ElevenLabs provider as shown above
  2. In a WhatsApp session, invoke the tts tool with any text
  3. Tool returns "Generated audio reply." (success)
  4. No audio file appears in /tmp/openclaw/tts-*/ or any media directory
  5. No audio is delivered to the WhatsApp chat

Expected Behavior

The TTS tool should generate an audio file and deliver it as a voice message to the WhatsApp chat.

Actual Behavior

  • Tool call returns success (Generated audio reply.)
  • No file is written to disk
  • No audio is sent to WhatsApp
  • No error in logs

Workaround

Using sag CLI (ElevenLabs TTS) to generate the file manually, then sending via message(action=send, filePath=..., asVoice=true) works correctly:

sag "text" --lang es -o /path/to/output.opus
# Then send via message tool with asVoice=true

This confirms:

  • ElevenLabs API key and config are valid
  • WhatsApp audio sending works
  • The issue is in the TTS tool's file generation/delivery pipeline

Additional Context

  • Reproduced consistently across multiple sessions and gateway restarts
  • First observed after upgrading to 2026.4.1, persists on 2026.4.2
  • The [[tts]] tag approach also fails silently

extent analysis

TL;DR

The TTS tool's failure to generate and deliver audio files to WhatsApp chats may be due to an issue in the file generation or delivery pipeline, which can be temporarily worked around by manually generating the audio file using the sag CLI and sending it via the message tool.

Guidance

  • Verify that the TTS tool is correctly configured to use the ElevenLabs provider and that the API key and configuration are valid, as confirmed by the successful manual generation and sending of audio files using the sag CLI.
  • Investigate the TTS tool's file generation pipeline to identify why no audio files are being written to disk, despite the tool returning a success message.
  • Check the permissions and access rights for the /tmp/openclaw/tts-*/ directory to ensure that the TTS tool has the necessary permissions to write files to this location.
  • Consider adding error logging or debugging statements to the TTS tool to help identify the root cause of the issue.

Example

No code snippet is provided as the issue is more related to configuration and pipeline issues rather than code-specific problems.

Notes

The issue seems to be specific to the TTS tool's integration with the ElevenLabs provider and WhatsApp, and may not be related to the underlying Node or OS versions. The fact that the sag CLI workaround works correctly suggests that the issue is isolated to the TTS tool's pipeline.

Recommendation

Apply workaround: Manually generate audio files using the sag CLI and send them via the message tool until the root cause of the TTS tool's issue is identified and fixed. This ensures that audio files can still be delivered to WhatsApp chats, albeit through a more manual process.

Vote matrix ยท 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?
Loadingโ€ฆ
#api#embedding generation#cache error#pipeline error#runtime error

Still need to ship something?

ร—6

Another batch ranked right after the header list โ€” different links, same matching logic.

Back to top recommendations

TRENDING