Text To Speech Print statements in Python SDK - 1.40.0

[boyko11]

After upgrading azure-cognitiveservices-speech to 1.40.0, in FastAPI python app I see PRINTS in the prod logs:
Info: on_underlying_io_bytes_received: Close frame received
Info: on_underlying_io_bytes_received: closing underlying io.
Info: on_underlying_io_close_complete: uws_state: 6.

2024-10-01 08:09:48,983 - INFO - [text_to_speech_service_azure.py:78] - Synthesizing sentence: I'm doing well, thank you for asking.
Info: on_underlying_io_bytes_received: Close frame received
Info: on_underlying_io_bytes_received: closing underlying io.
Info: on_underlying_io_close_complete: uws_state: 6.
2024-10-01 08:09:49,567 - INFO - [text_to_speech_service_azure.py:78] - Synthesizing sentence: It's always nice to start the day with a friendly chat.
Info: on_underlying_io_bytes_received: Close frame received
Info: on_underlying_io_bytes_received: closing underlying io.
Info: on_underlying_io_close_complete: uws_state: 6.
2024-10-01 08:09:50,284 - INFO - [llm_service_abstract.py:77] - entire_llm_response: I'm doing well, thank you for asking. It's always nice to start the day with a friendly chat. What's on your agenda for today?
2024-10-01 08:09:50,286 - INFO - [text_to_speech_service_azure.py:78] - Synthesizing sentence: What's on your agenda for today?
Info: on_underlying_io_bytes_received: Close frame received
Info: on_underlying_io_bytes_received: closing underlying io.
Info: on_underlying_io_close_complete: uws_state: 6.

Using text-to-speech on a python web app:

    async def text_to_audio_stream(
        self, text: str, user_id: str
    ) -> AsyncGenerator[bytes, None]:
        await TimingUtil().enqueue_timing_event(
            TimingEventName.tts_sentence_received,
            user_id,
            time.time(),
            {"sentence": text},
        )

        logger.info("Synthesizing sentence: %s", text)
        synthesizer = speechsdk.SpeechSynthesizer(
            speech_config=self.speech_config, audio_config=None
        )

        ssml = self.ssml_string_template.format(
            voice_name=self.azure_config.speech_synthesis_voice_name,
            voice_speed_increase=self.azure_config.speech_synthesis_voice_speed_increase,
            text=text,
        )

        logger.debug(ssml)

        result = await asyncio.get_running_loop().run_in_executor(
            None,
            lambda _synthesizer=synthesizer, _ssml=ssml: _synthesizer.speak_ssml_async(  # type: ignore
                _ssml
            ).get(),
        )

        if result.reason == speechsdk.ResultReason.SynthesizingAudioCompleted:
            await TimingUtil().enqueue_timing_event(
                TimingEventName.tts_sentence_synthesized, user_id, time.time()
            )

            audio_data_stream = speechsdk.AudioDataStream(result)
            buffer_size = 4096

            while True:
                audio_bytes_buffer = bytes(buffer_size)
                audio_bytes_read = audio_data_stream.read_data(audio_bytes_buffer)
                logger.debug("%s bytes received.", audio_bytes_read)
                if audio_bytes_read == 0:
                    break
                yield audio_bytes_buffer[:audio_bytes_read]

        else:
            logger.warning("Speech synthesis possible fail: %s", result.reason)

After downgrading back to azure-cognitiveservices-speech = "1.37.0"
the prints disappeared:

2024-10-01 08:27:28,744 - INFO - [text_to_speech_service_azure.py:78] - Synthesizing sentence: Good to hear everything's under control, Boyko.
2024-10-01 08:27:29,481 - INFO - [text_to_speech_service_azure.py:78] - Synthesizing sentence: As for me, I'm here and ready to chat as always.
2024-10-01 08:27:30,225 - INFO - [llm_service_abstract.py:77] - entire_llm_response: Good to hear everything's under control, Boyko. As for me, I'm here and ready to chat as always. What would you like to talk about today?
2024-10-01 08:27:30,227 - INFO - [text_to_speech_service_azure.py:78] - Synthesizing sentence: What would you like to talk about today?

Could you please remove the prints or switch them over to logging? Could you also add a linter that detects prints before you push the code as a package. Thanks!

[boyko11]

The prints started appearing in 1.38.0:

[github-actions]

This item has been open without activity for 19 days. Provide a comment on status and remove "update needed" label.

[boyko11]

Tried the latest this morning - 1.41.1. The prints(or log statements) are still there. Please change them over to debug, or silence them in some other way so we can upgrade. Thanks!

[ShadowStrikerJ]

Encountering the same issue with the speech to text recognizer, can't find any way to remove the logging statements

[yash-sprnklr]

Facing the same issue!