What causes "failed to request ElevenLabs music: {exc}" in harry0703/MoneyPrinterTurbo?

Retry the generation once or twice with backoff — the error is explicitly treated as transient degradation, not a hard failure

harry0703/MoneyPrinterTurbo · error · ElevenLabsMusicError

failed to request ElevenLabs music: {exc}

Error message

failed to request ElevenLabs music: {exc}

What it means

Wraps any requests.RequestException (DNS failure, connection reset, TLS error, read timeout from the (15, <=1800s) timeout tuple, or a dropped stream mid-download) raised while POSTing to ElevenLabs or streaming the audio body back. Wrapping it in ElevenLabsMusicError forces the task into its degradation logic instead of failing an already-rendered video because of third-party network churn.

Source

Thrown at app/services/elevenlabs_music.py:342

                            (Path(video_path).name, video_file, "video/mp4"),
                        )
                    ],
                    data=request_data,
                    stream=True,
                    timeout=_request_timeout(),
                )
                with response:
                    if not response.ok:
                        raise ElevenLabsMusicError(
                            "ElevenLabs generation failed "
                            f"({response.status_code}): "
                            f"{_safe_response_error(response)}"
                        )
                    total_bytes = _stream_audio(response, temp_audio_path)
        except requests.RequestException as exc:
            # 下载阶段断线也属于请求失败，必须进入任务降级逻辑，不能留下半条
            # 音频或让已经生成的视频因为第三方网络波动整体失败。
            raise ElevenLabsMusicError(
                f"failed to request ElevenLabs music: {exc}"
            ) from exc

        try:
            bgm_service.validate_audio_file(temp_audio_path, timeout_seconds=120)
        except (bgm_service.BgmUploadError, bgm_service.BgmServiceError) as exc:
            raise ElevenLabsMusicError(
                "ElevenLabs returned audio that FFmpeg cannot decode"
            ) from exc
        os.replace(temp_audio_path, output_path)
        temp_audio_path = ""
        logger.info(
            "ElevenLabs background music generated: "
            f"output={output_path}, size={total_bytes} bytes"
        )
        return output_path
    finally:
        _remove_file(temp_audio_path)

View on GitHub (pinned to 1f9f19c202)

Solutions

Retry the generation once or twice with backoff — the error is explicitly treated as transient degradation, not a hard failure
Increase the elevenlabs.music_timeout config value if the read timeout fires on long videos (it is capped at 1800s)
Verify network egress to api.elevenlabs.io (curl -I https://api.elevenlabs.io) and fix proxy/DNS if the connection never opens
If it fails repeatedly, let the task fall back to no-BGM output rather than aborting the rendered video

Example fix

# before
path = _request_bgm(proxy_path, output_path, prompt)

# after
for attempt in range(2):
    try:
        path = _request_bgm(proxy_path, output_path, prompt)
        break
    except ElevenLabsMusicError as exc:
        if attempt == 1:
            raise
        time.sleep(10)

Defensive patterns

Strategy: retry

Validate before calling

import socket, urllib.parse
host = urllib.parse.urlparse("https://api.elevenlabs.io").hostname
socket.gethostbyname(host)  # fail fast on DNS problems

Try / catch

for attempt in range(3):
    try:
        return elm.generate_bgm(video, out, duration, prompt)
    except elm.ElevenLabsMusicError as e:
        if attempt == 2 or not is_transient(e):
            return fallback_no_bgm()
        time.sleep(2 ** attempt * 5)

Prevention

Size elevenlabs.music_timeout generously for long videos (cap 1800s)
Run generation with retry + exponential backoff wrapped around generate_bgm
Degrade to no-BGM output on repeated network failure rather than failing the video

When it happens

Trigger: Any requests exception during the streaming POST /v1/music/video-to-music or during _stream_audio: ConnectionError (no route to api.elevenlabs.io), ChunkedEncodingError (server closed the stream mid-audio), ReadTimeout when generation exceeds the configured music_timeout read timeout.

Common situations: Corporate proxies/firewalls blocking api.elevenlabs.io, flaky home uplinks dropping long streams, music_timeout set too low for long videos (600s content can take minutes), DNS failures in containers.

Related errors

AI-assisted analysis of harry0703/MoneyPrinterTurbo@1f9f19c202 (2026-08-14). Data as JSON: /api/errors/16d23dec395a3ec3. Report an issue: GitHub.