"We could not parse the JSON body of your request." with tts-1-(hd) endpoint, text to speach

[zopyx]

Confirm this is an issue with the Python library and not an underlying OpenAI API

• This is an issue with the Python library

Describe the bug

I am receiving this error (suddenly) with openai==1.12.0 on Feb 09, 2024...worked the day before:

  File "/Users/ajung/Dropbox/FILMORA- VIDEOS/Typesense-Onkopedia - 2/convert.py", line 66, in <module>
    text_to_speech(content, "de", output_wav)
  File "/Users/ajung/Dropbox/FILMORA- VIDEOS/Typesense-Onkopedia - 2/convert.py", line 17, in text_to_speech
    response = client.audio.speech.create(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/ajung/Dropbox/FILMORA- VIDEOS/Typesense-Onkopedia - 2/lib/python3.12/site-packages/openai/resources/audio/speech.py", line 81, in create
    return self._post(
           ^^^^^^^^^^^
  File "/Users/ajung/Dropbox/FILMORA- VIDEOS/Typesense-Onkopedia - 2/lib/python3.12/site-packages/openai/_base_client.py", line 1201, in post
    return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls))
                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/ajung/Dropbox/FILMORA- VIDEOS/Typesense-Onkopedia - 2/lib/python3.12/site-packages/openai/_base_client.py", line 889, in request
    return self._request(
           ^^^^^^^^^^^^^^
  File "/Users/ajung/Dropbox/FILMORA- VIDEOS/Typesense-Onkopedia - 2/lib/python3.12/site-packages/openai/_base_client.py", line 981, in _request
    raise self._make_status_error_from_response(err.response) from None
openai.BadRequestError: Error code: 400 - {'error': {'message': "We could not parse the JSON body of your request. (HINT: This likely means you aren't using your HTTP library correctly. The OpenAI API expects a JSON payload, but what was sent was not valid JSON. If you have trouble figuring out how to fix this, please contact us through our help center at help.openai.com.)", 'type': 'invalid_request_error', 'param': None, 'code': None}}

To Reproduce

import sys
from openai import OpenAI
import openai

client = OpenAI()

SPEED = 1.0
VOICE = "echo"
MODEL="tts-1",
#MODEL="tts-1-hd",

def text_to_speech(content: str, language: str, output_wav: str):

    response = client.audio.speech.create(
        model=MODEL,
        voice=VOICE,
        speed=SPEED,
        input=content,
    )

    print(f"Saved to {output_wav}")
    response.stream_to_file(output_wav)


text_to_speech("hello world", "de", "output.wav")

Code snippets

No response

OS

MacOS Soama

Python version

3.12

Library version

open 1.12.0

[rattrayalex]

Does this happen with both tts-1 and tts-1-hd models? Does the same problem happen if you use the same content you were sending a few days ago?

[zopyx]

Invalid issue. My code defines MODEL as a one-element tuple instead of a string (copy&paste refactoring issue).