Azure Text-to Speech Error Code: 0x38 (SPXERR_AUDIO_SYS_LIBRARY_NOT_FOUND) when using Python 3.10 on Wep App Service

Question

I followed these 2 guides and everything works fine on my local windows 10 machine using python 3.10., but when I deploy my code using VS Code to my linux basic web app service, running python 3.10, I receive the following error.

Guides:

https://learn.microsoft.com/en-us/azure/cognitive-services/speech-service/quickstarts/setup-platform?tabs=linux%2Cubuntu%2Cdotnet%2Cjre%2Cmaven%2Cnodejs%2Cmac%2Cpypi&pivots=programming-language-python

and
https://learn.microsoft.com/en-us/azure/cognitive-services/speech-service/get-started-text-to-speech?tabs=linux%2Cterminal&pivots=programming-language-python

Troubleshooting:

https://learn.microsoft.com/en-us/azure/cognitive-services/speech-service/troubleshooting?tabs=powershell#spxerr_audio_sys_library_not_found

Code:

def response_to_speech(response):

    subscription_key = os.getenv("AZURE_SPEECH")
    region = os.getenv("AZURE_REGION")

    speech_config = speechsdk.SpeechConfig(subscription=subscription_key, region=region)
    audio_config = speechsdk.audio.AudioOutputConfig(use_default_speaker=True)
    speech_config.speech_synthesis_voice_name='en-US-JennyNeural'

    speech_synthesizer = speechsdk.SpeechSynthesizer(speech_config=speech_config, audio_config=audio_config)

    audio = speech_synthesizer.speak_text_async(response).get()

    return audio

Dependencies:

Python 3.10.9

libasound2 is already the newest version (1.2.4-1.1).
build-essential is already the newest version (12.9).
libssl-dev is already the newest version (1.1.1n-0+deb11u3).
wget is already the newest version (1.21-1+deb11u1).

Requirement already satisfied: azure-cognitiveservices-speech in ./antenv/lib/python3.10/site-packages (1.25.0)

Error:

Exception Value: Exception with error code: 
[CALL STACK BEGIN]
/tmp/8db182897509f5b/antenv/lib/python3.10/site-packages/azure/cognitiveservices/speech/libMicrosoft.CognitiveServices.Speech.core.so(+0x185fa4) [0x7f1b1a187fa4]
/tmp/8db182897509f5b/antenv/lib/python3.10/site-packages/azure/cognitiveservices/speech/libMicrosoft.CognitiveServices.Speech.core.so(+0x1ee61a) [0x7f1b1a1f061a]
/tmp/8db182897509f5b/antenv/lib/python3.10/site-packages/azure/cognitiveservices/speech/libMicrosoft.CognitiveServices.Speech.core.so(synthesizer_create_speech_synthesizer_from_config+0xf3) [0x7f1b1a09c9d0]
/usr/lib/x86_64-linux-gnu/libffi.so.7(+0x6d1d) [0x7f1b35120d1d]
/usr/lib/x86_64-linux-gnu/libffi.so.7(+0x6289) [0x7f1b35120289]
/opt/python/3.10.9/lib/python3.10/lib-dynload/_ctypes.cpython-310-x86_64-linux-gnu.so(+0x11c19) [0x7f1b35137c19]
/opt/python/3.10.9/lib/python3.10/lib-dynload/_ctypes.cpython-310-x86_64-linux-gnu.so(+0x11493) [0x7f1b35137493]
/opt/python/3.10.9/lib/libpython3.10.so.1.0(PyObject_Call+0xc3) [0x7f1b38e0d333]
/opt/python/3.10.9/lib/libpython3.10.so.1.0(_PyEval_EvalFrameDefault+0x2a4b) [0x7f1b38df148b]
/opt/python/3.10.9/lib/libpython3.10.so.1.0(_PyFunction_Vectorcall+0x78) [0x7f1b38e00218]
/opt/python/3.10.9/lib/libpython3.10.so.1.0(PyObject_Call+0x11a) [0x7f1b38e0d38a]
/opt/python/3.10.9/lib/libpython3.10.so.1.0(_PyEval_EvalFrameDefault+0x2a4b) [0x7f1b38df148b]
/opt/python/3.10.9/lib/libpython3.10.so.1.0(_PyFunction_Vectorcall+0x78) [0x7f1b38e00218]
/opt/python/3.10.9/lib/libpython3.10.so.1.0(_PyObject_FastCallDictTstate+0x16d) [0x7f1b38df82dd]
/opt/python/3.10.9/lib/libpython3.10.so.1.0(+0x15f5e5) [0x7f1b38e095e5]
/opt/python/3.10.9/lib/libpython3.10.so.1.0(_PyObject_MakeTpCall+0x273) [0x7f1b38df9143]
/opt/python/3.10.9/lib/libpython3.10.so.1.0(_PyEval_EvalFrameDefault+0x6079) [0x7f1b38df4ab9]
[CALL STACK END]
Exception with an error code: 0x38 (SPXERR_AUDIO_SYS_LIBRARY_NOT_FOUND)

Traceback:

Traceback (most recent call last):
  File "/tmp/8db182897509f5b/antenv/lib/python3.10/site-packages/django/core/handlers/exception.py", line 56, in inner
    response = get_response(request)
  File "/tmp/8db182897509f5b/antenv/lib/python3.10/site-packages/django/core/handlers/base.py", line 197, in _get_response
    response = wrapped_callback(request, *callback_args, **callback_kwargs)
  File "/tmp/8db182897509f5b/antenv/lib/python3.10/site-packages/django/contrib/auth/decorators.py", line 23, in _wrapped_view
    return view_func(request, *args, **kwargs)
  File "/tmp/8db182897509f5b/src/views/chat.py", line 85, in as_view
    audio_response = response_to_speech(text_response)
  File "/tmp/8db182897509f5b/src/views/chat.py", line 129, in response_to_speech
    speech_synthesizer = speechsdk.SpeechSynthesizer(speech_config=speech_config, audio_config=audio_config)
  File "/tmp/8db182897509f5b/antenv/lib/python3.10/site-packages/azure/cognitiveservices/speech/speech.py", line 2147, in __init__
    _call_hr_fn(fn=_sdk_lib.synthesizer_create_speech_synthesizer_from_config, *[
  File "/tmp/8db182897509f5b/antenv/lib/python3.10/site-packages/azure/cognitiveservices/speech/interop.py", line 62, in _call_hr_fn
    _raise_if_failed(hr)
  File "/tmp/8db182897509f5b/antenv/lib/python3.10/site-packages/azure/cognitiveservices/speech/interop.py", line 55, in _raise_if_failed
    __try_get_error(_spx_handle(hr))
  File "/tmp/8db182897509f5b/antenv/lib/python3.10/site-packages/azure/cognitiveservices/speech/interop.py", line 50, in __try_get_error
    raise RuntimeError(message)

Answer 1

Jared Rice I think on the remote app service the default audio config needs to be set to an audio file instead of default as in local machine it cannot default to a speaker in this case.

Try adding the following to update audio_config

    file_name = "outputaudio.wav"
    file_config = speechsdk.audio.AudioOutputConfig(filename=file_name)
    speech_synthesizer = speechsdk.SpeechSynthesizer(speech_config=speech_config, audio_config=file_config)

Answer 2

file_name = "outputaudio.wav"your text

Answer 3

I get a (SPXERR_AUDIO_SYS_LIBRARY_NOT_FOUND) error when I stream audio on Azure. I switched to streaming because the Azure server did not like me writing a file.

So my code works with streaming (or writing a file) locally , but it fails on the Azure server. What is the SPXERR_AUDIO_SYS_LIBRARY_NOT_FOUND error that I'm getting from the server?

Is there a way to get help from someone at Azure?

2025-03-07T18:58:12.4613574Z /opt/python/3.12.8/lib/libpython3.12.so.1.0(+0x24b859) [0x7cbe564fb859]

2025-03-07T18:58:12.4613639Z /opt/python/3.12.8/lib/libpython3.12.so.1.0(_PyObject_MakeTpCall+0x2cb) [0x7cbe564b992b]

2025-03-07T18:58:12.4613659Z /opt/python/3.12.8/lib/libpython3.12.so.1.0(_PyEval_EvalFrameDefault+0x713) [0x7cbe564c2353]

2025-03-07T18:58:12.4613676Z /opt/python/3.12.8/lib/libpython3.12.so.1.0(+0x11399a) [0x7cbe563c399a]

2025-03-07T18:58:12.4613693Z /opt/python/3.12.8/lib/libpython3.12.so.1.0(_PyObject_FastCallDictTstate+0x1d5) [0x7cbe564bc3a5]

2025-03-07T18:58:12.4613911Z [CALL STACK END]

2025-03-07T18:58:12.4613954Z

2025-03-07T18:58:12.4613978Z Exception with an error code: 0x38 (SPXERR_AUDIO_SYS_LIBRARY_NOT_FOUND)

2025-03-07T18:58:12.4614016Z 169.254.129.1 - - [07/Mar/2025:18:58:12 +0000] "GET /synthesize?text=Hello%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20&voice=en-US-AvaMultilingualNeural HTTP/1.1" 500 33 "https://flasktranslatorapp-bvepdbcvgmh7ffek.westus2-01.azurewebsites.net/" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/133.0.0.0 Safari/537.36"

2025-03-07T18:59:57.6958825Z 169.254.129.1 - - [07/Mar/2025:18:59:57 +0000] "GET /admin/host/status HTTP/1.1" 404 207 "-" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/133.0.0.0 Safari/537.36"Ending Log Tail of existing logs ---Starting Live Log Stream ---