Riva Embedded 2.7 : ASR Mic scripts not working

AI & Data Science Deep Learning (Training & Inference) Riva
,
#1

I have Riva 2.7 version installed on jetson xavier nx.
I am trying to run transcribe_mic.py script but failed with the below error-

Traceback (most recent call last):
File “transcribe_mic.py”, line 75, in
main()
File “transcribe_mic.py”, line 60, in main
with riva.client.audio_io.MicrophoneStream(
File “/usr/local/lib/python3.8/dist-packages/riva/client/audio_io.py”, line 24, in enter
self._audio_stream = self._audio_interface.open(
File “/usr/lib/python3/dist-packages/pyaudio.py”, line 750, in open
stream = Stream(self, *args, **kwargs)
File “/usr/lib/python3/dist-packages/pyaudio.py”, line 441, in init
self._stream = pa.open(**arguments)
OSError: [Errno -9997] Invalid sample rate

Please provide the following information when requesting support.

Hardware - GPU (Nvidia Jetson Xavier NX)
Hardware - CPU
Operating System Jetpack 5.0.2
Riva Version 2.7.0

I raised this issue 4 days ago on Github.
Looking forward.

@SunilJB

#2

Hi @user115721

Thanks for your interest in Riva

I will check regarding this error with the team and get back to you,

Thanks

#3

Hi @user115721

Thanks for waiting for us,

I have updates from the team,

The python ASR client app has a limitation that it works only for headsets that support specific capabilities in terms of sample rate etc. Logitech H340 works

So, the python client app may not be able to work depending on the headset used because of this limitation.

Recommendation is to use C++ client app Command-line Clients — NVIDIA Riva with the “–audio_device” argument for mic input (see --help for more detail).

This should support all headsets when “plughw:x,y” device_id is provided, since that does internal sample rate conversion

Thanks