created windows audio client #38
Conversation
| import pyaudiowpatch as pyaudio | ||
|
|
||
| def translator(): | ||
| DOCKER_IP = "http://0.0.0.0:8000/" |
There was a problem hiding this comment.
This should be a global or a parameter passed into the translator function because the IP address of the server is not static and may need to be changed by the user.
Consider allowing the user to pass in the addr/port as command line arguments.
|
|
||
| def translator(): | ||
| DOCKER_IP = "http://0.0.0.0:8000/" | ||
| print("TRANSLATOR IS RUNNING") |
There was a problem hiding this comment.
Bad practice to use print statements like this. Use the logging module instead
| # open PyAudio manager via context manager | ||
| with pyaudio.PyAudio() as p: | ||
| # open audio stream via context manager | ||
| with p.open(format=pyaudio.paInt32, channels=2, rate=48000, input=True, frames_per_buffer=1024) as stream: |
There was a problem hiding this comment.
audio settings change depending on the user's setup, it would be nice to not have these be hard-coded values
Also, the input device is never specified. Look at the examples in the PyAudioWPatch repo for how you are supposed to use this.
| } | ||
|
|
||
| # send audio data to Docker container | ||
| with requests.Session() as session: |
There was a problem hiding this comment.
having a session that is immediately closed defeats the purpose of the session. It would be better to either use requests.post or to pass the session into this function so each post request can use the same session.
| "avg": float(avg), | ||
| "peak": float(peak), | ||
| "data": data.tolist(), | ||
| "source": "Windows Device" |
There was a problem hiding this comment.
The source tag is supposed to refer to which audio input is being used - saying 'windows device' is fine but it would be more helpful to grab the actual name/id of the input device that is getting recorded.
|
closes #30 |
BCaven
changed the title
2 participants
No description provided.