Crash_Override/FreeDATA
1
0
Fork 0
mirror of https://github.com/DJ2LS/FreeDATA synced 2024-05-14 08:04:33 +00:00
Code Issues Packages Projects Releases Wiki Activity

catching exceptions on stream.read() to help us spot buffer size issues, also saving a file of samples received to help debugging

Browse source
This commit is contained in:
drowe67 2021-12-16 17:52:42 +10:30 committed by David Rowe
parent 27a43f1cb0
commit fa4e22c64b
1 changed files with 24 additions and 11 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

35
test/001_highsnr_stdio_audio/test_rx.py
View file

@ -47,7 +47,7 @@ DEBUGGING_MODE = args.DEBUGGING_MODE
TIMEOUT = args.TIMEOUT
# AUDIO PARAMETERS
AUDIO_FRAMES_PER_BUFFER = 2400 # seems to be best if >=1024
AUDIO_FRAMES_PER_BUFFER = 2400*2 # <- consider increasing if you get nread_exceptions > 0
MODEM_SAMPLE_RATE = codec2.api.FREEDV_FS_8000
AUDIO_SAMPLE_RATE_RX = 48000
@ -101,24 +101,33 @@ rx_total_frames = 0
rx_frames = 0
rx_bursts = 0
rx_errors = 0
nread_exceptions = 0
timeout = time.time() + TIMEOUT
receive = True
audio_buffer = codec2.audio_buffer(codec2.api.freedv_get_n_max_modem_samples(freedv))
AUDIO_FRAMES_PER_BUFFER_8K = int(AUDIO_FRAMES_PER_BUFFER/codec2.api.FDMDV_OS_48)
audio_buffer = codec2.audio_buffer(AUDIO_FRAMES_PER_BUFFER*2)
resampler = codec2.resampler()
# Copy received 48 kHz to a file. Listen to this file with:
# aplay -r 48000 -f S16_LE rx48.raw
# Corruption of this file is a good way to detect audio card issues
frx = open("rx48.raw", mode='wb')
# initial number of samples we need
nin = codec2.api.freedv_nin(freedv)
while receive and time.time() < timeout:
if AUDIO_INPUT_DEVICE != -1:
data_in48k = stream_rx.read(AUDIO_FRAMES_PER_BUFFER, exception_on_overflow = False)
if AUDIO_INPUT_DEVICE != -1:
try:
data_in48k = stream_rx.read(AUDIO_FRAMES_PER_BUFFER, exception_on_overflow = True)
except OSError as err:
print(err, file=sys.stderr)
nread_exceptions += 1
else:
data_in48k = sys.stdin.buffer.read(AUDIO_FRAMES_PER_BUFFER*2)
# insert samples in buffer
x = np.frombuffer(data_in48k, dtype=np.int16)
x.tofile(frx)
if len(x) != AUDIO_FRAMES_PER_BUFFER:
receive = False
x = resampler.resample48_to_8(x)
@ -154,12 +163,16 @@ while receive and time.time() < timeout:
rx_bursts = rx_bursts + 1
if rx_bursts == N_BURSTS:
receive = False
receive = False
if time.time() >= timeout:
print("TIMEOUT REACHED")
if nread_exceptions:
print("nread_exceptions %d - receive audio lost! Consider increasing Pyaudio frames_per_buffer..." % \
nread_exceptions, file=sys.stderr)
print(f"RECEIVED BURSTS: {rx_bursts} RECEIVED FRAMES: {rx_total_frames} RX_ERRORS: {rx_errors}", file=sys.stderr)
frx.close()
# and at last check if we had an openend pyaudio instance and close it
if AUDIO_INPUT_DEVICE != -1: