Streaming from microphone help

[S2741940]

I'm working through the Live Streaming Starter Kit, and am encountering an error when running the command: python3.9 test_suite.py -k YOUR_DEEPGRAM_API_KEY -i mic

ERROR
ℹ️ Request ID: 0781a4ed-5353-436b-ac55-a7927e0a8311
ℹ️ Model: general
🟢 (1/5) Successfully opened Deepgram streaming connection
🟢 (2/5) Ready to stream mic audio to Deepgram. Speak into your microphone to transcribe.
Error while sending: Task <Task pending name='Task-5' coro=<run..sender() running at ~/test_suite.py:101> cb=[_gather.._done_callback() at /usr/local/Cellar/[email protected]/3.9.18_2/Frameworks/Python.framework/Versions/3.9/lib/python3.9/asyncio/tasks.py:767]> got Future attached to a different loop
🔴 ERROR: Something went wrong! Task <Task pending name='Task-5' coro=<run..sender() running at ~/test_suite.py:101> cb=[_gather.._done_callback() at /usr/local/Cellar/[email protected]/3.9.18_2/Frameworks/Python.framework/Versions/3.9/lib/python3.9/asyncio/tasks.py:767]> got Future attached to a different loop

Mac OS12.3 with Mac Studio M1 Max

I've read that there can be issues with Python versions. When I try to run the same in a Python3.10 or 3.12 environment, I get:
Could not import the PyAudio C module 'pyaudio._portaudio'.
Traceback (most recent call last):
File "~/test_suite.py", line 1, in
import pyaudio
File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/pyaudio/init.py", line 111, in
import pyaudio._portaudio as pa
ImportError: dlopen(/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/pyaudio/_portaudio.cpython-312-darwin.so, 0x0002): symbol not found in flat namespace '_PaMacCore_SetupChannelMap'

I am able to successfully run the 'stream a file' and 'stream my own WAV' commands when using Python3.9.

Any help would be much appreciated, thanks!

For my 3.9 environment:

Package Version

---

aiohttp 3.9.3
aiosignal 1.3.1
anyio 4.2.0
async-timeout 4.0.3
attrs 23.2.0
certifi 2024.2.2
chardet 4.0.0
dataclasses-json 0.6.4
deepgram-sdk 3.1.3
exceptiongroup 1.2.0
frozenlist 1.4.1
h11 0.14.0
httpcore 1.0.2
httpx 0.26.0
idna 3.6
marshmallow 3.20.2
multidict 6.0.5
mypy-extensions 1.0.0
packaging 23.2
pip 24.0
PyAudio 0.2.13
pydub 0.25.1
setuptools 68.2.2
sniffio 1.3.0
typing_extensions 4.9.0
typing-inspect 0.9.0
verboselogs 1.7
websockets 12.0
yarl 1.9.4

[team-deepgram]

Thanks for asking your question about Deepgram! If you didn't already include it in your post, please be sure to add as much detail as possible so we can assist you efficiently, such as:

• The request_id if you have a question about your requests or transcription responses.
• The features you used or the full api.deepgram.com URL you sent your request to, including parameters.
• Any code snippets you can share.

[davidvonthenen]

Hi @S2741940

You need to install the portaudio library on your system.

[S2741940]

Hi, thanks for your reply.

I already have portaudio installed. If I go: brew portaudio install, I get a message saying 'Warning: portaudio 19.7.0 is already installed and up-to-date.' I just reinstalled an no change.

[davidvonthenen]

Yea, we don't support Python 3.9. The minimum version is 3.10.

Linux isn't able to find the library based on the error message here:

ImportError: dlopen(/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/pyaudio/_portaudio.cpython-312-darwin.so, 0x0002): symbol not found in flat namespace '_PaMacCore_SetupChannelMap'

After installing, did you reboot your OS (or ldconfig the system)?

[alexanderkoller]

Yea, we don't support Python 3.9. The minimum version is 3.10.

This is a really important piece of information. Could you document it somewhere on the website so future Python users won't fall into the same trap as the OP and me?

[davidvonthenen]

There is a PyAudio==0.2.14 you might want to update to. Also want to make sure you are using Python from like brew and not using some apple version of it.

[S2741940]

Thanks for your help and patience!

I've tried the above and still no luck unfortunately. I am able to navigate to the _portaudio.cpython-312-darwin.so file so it does exist.

Here are my steps attempting to do the basic test in Python 3.10 (Mac OSX 12.3):

MAKE NEW ENVIRONMENT:
python3.10 -m venv myenvthreeten

ENTER ENVIRONEMNT:
source myenvthreeten/bin/activate

INSTALL PORT AUDIO
brew install portaudio

INSTALL REQUIREMENTS:
pip3.10 install -r requirements.txt

UPDATE PIP
/Users/XXX/myenvthreeten/bin/python3.10 -m pip install --upgrade pip

UPDATE PYAUDIO
pip install pyaudio==0.2.14

RESTART COMPUTER

ENTER ENVIRONEMNT:
source myenvthreeten/bin/activate

RUN SCRIPT
python3.10 test_suite.py -k [API KEY}

I receive this error:
Could not import the PyAudio C module 'pyaudio._portaudio'.
Traceback (most recent call last):
File "/Users/XXX/test_suite.py", line 1, in
import pyaudio
File "/Users/XXX/myenvthreeten/lib/python3.10/site-packages/pyaudio/init.py", line 111, in
import pyaudio._portaudio as pa
ImportError: dlopen(/Users/XXX/myenvthreeten/lib/python3.10/site-packages/pyaudio/_portaudio.cpython-310-darwin.so, 0x0002): symbol not found in flat namespace '_PaMacCore_SetupChannelMap'

I've tried installing and reinstalling portaudio and pyaudio a bunch of times, and also installing port audio from the website download and performing the ./configure && make command. I can run the python3.9 test_suite.py -k [API KEY} command in the same environment with 3.9 installed and it works, just doesn't work at all in 3.10.

After this I have the following pip list:
Package Version

---

aiohttp 3.7.4.post0
async-timeout 3.0.1
attrs 23.2.0
chardet 4.0.0
idna 3.6
multidict 6.0.5
pip 24.0
PyAudio 0.2.14
pydub 0.25.1
setuptools 57.4.0
typing_extensions 4.9.0
websockets 10.3
yarl 1.9.4

Any ideas would be great, all good if I'm just beyond help! Thank you

[davidvonthenen]

I should have asked this upfront initially, but do you have Rosetta installed?

[S2741940]

Yes, Rosetta is installed. On 10 Feb 2024, at 1:59 am, David vonThenen ***@***.***> wrote:  I should have asked this upfront initially, but do you have Rosetta installed? — Reply to this email directly, view it on GitHub<#592 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/BGAFFTWCYEJ266XKWBLZ723YSZBW3AVCNFSM6AAAAABC7B2OZWVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4DIMRQGY3DI>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[davidvonthenen]

I am looking back at your posts and there seem to be a bunch of inconsistencies regarding the the environment.

You are using python3.10 as your python version and yet there seems to be grabbing _portaudio.cpython-312-darwin.so? This is super odd.

[S2741940]

Hey, thanks for your continued help!

I think those inconsistencies are just me giving examples from when I try the process using python3.12 (I've tried both 3.12 and 3.10)

I just followed my steps above and was greeted with the following error:
Could not import the PyAudio C module 'pyaudio._portaudio'.
Traceback (most recent call last):
File "/Users/XXX/test_suite.py", line 1, in
import pyaudio
File "/Users/XXX/myenvthreetentest/lib/python3.10/site-packages/pyaudio/init.py", line 111, in
import pyaudio._portaudio as pa
ImportError: dlopen(/Users/XXX/myenvthreetentest/lib/python3.10/site-packages/pyaudio/_portaudio.cpython-310-darwin.so, 0x0002): symbol not found in flat namespace '_PaMacCore_SetupChannelMap'

[davidvonthenen]

My only other suggestion now at this point would be to try and compile portaudio yourself to find out what symbols in the so are missing. that should point you in the direction of what dependencies are missing.