ElevenLab speech to speech "Missing boundary in multipart."

How To
,
1

Hi everyone!

I’m facing an issue while trying to set up an HTTP request in Make.com to work with the ElevenLabs Speech-to-Speech API. I keep running into this error:
{"detail":"Missing boundary in multipart."}

Here’s what I’m doing:

  • URL: https://api.elevenlabs.io/v1/speech-to-speech/XqPohkT4dEpjPmwoE1j8
  • Headers:
    • Content-Type: multipart/form-data; boundary=----MakeBoundary12345
    • xi-api-key: [my API key]
  • Request Body:

css

Копировать код

----MakeBoundary12345
Content-Disposition: form-data; name="audio_file"; filename="audio.wav"
Content-Type: audio/wav

[Binary data of the audio file here]
----MakeBoundary12345
Content-Disposition: form-data; name="model_id"

eleven_english_sts_v2
----MakeBoundary12345
Content-Disposition: form-data; name="settings"

{"stability":0.5,"similarity_boost":0.75,"pitch_shift":1.2,"speed_up":1.1}
----MakeBoundary12345
Content-Disposition: form-data; name="output_format"

mp3_44100_128
----MakeBoundary12345
Content-Disposition: form-data; name="remove_background_noise"

true
----MakeBoundary12345--

I’m pulling the audio file from Google Drive, converting it into binary, and including it in the request body. But the API keeps throwing the “missing boundary” error.

Make support told me that the issue lies with ElevenLabs, but I feel like there’s something off in how I’m configuring the request itself.

Has anyone else dealt with this? I’d really appreciate any tips or examples on how to configure the request properly so it works. Any help would mean a lot!

Thanks in advance!

Best regards,
Sergey