Python valueerror invalid literal for int with base 10

[JRMeyer]

>>> Toaster1337
[December 8, 2020, 2:09pm]

I'm record and save it to an audio variable and got this error:

return _impl.SpeechToText(aCtx, aBuffer)
ValueError: invalid literal for int() with base 10: b'# slash x06 slash x0e slash x0b slash x92 slash n0 slash x11 slash xac slash x18 slash xc2 slash x1b slash x05 slash x1c slash x87 slash x1d slash xd1 slash x18 slash xd0 slash x12v slash x07 slash xc6 slash xf9 slash xf1 slash xf0 slash x8c slash xe7[ slash xe5 slash x12 slash xe7$ slash xe5 slash x90 slash xe3 slash xe7 slash x14 slash xef slash xdf slash xf2 slash xbe slash xf0 slash xbb slash xefy slash xeb slash x93 slash xe49 slash xe0 slash x05 slash xde slash xd6 slash xd9

What do it means and how can i fix it?

def run(self, audio): slash
slash # audio = BytesIO(audio) slash
result = deepspeechModel.stt(audio_buffer=audio) slash
return result

[This is an archived TTS discussion thread from discourse.mozilla.org/t/python-valueerror-invalid-literal-for-int-with-base-10]

[JRMeyer]

>>> othiele
[December 8, 2020, 2:06pm]

It is hard to help with just 2 lines of code, please give more
information.

[Archived Post]

[JRMeyer]

>>> Toaster1337
[December 8, 2020, 2:17pm]

Recorded with Pyaudio:

FORMAT = pyaudio.paInt16 slash
CHANNELS = 1 slash
RATE = 16000 slash
CHUNK = 1024 slash
rec = slash [ slash ]

pyAudio.open(rate=RATE, slash
channels=CHANNELS, slash
format=FORMAT, slash
input=True, slash
output=True, slash
frames_per_buffer=CHUNK)

data = self. slash _stream.read(CHUNK) slash
rec.append(data) slash
run(b''.join(rec))

The error is in line: slash
result = deepspeechModel.stt(audio_buffer=audio)

[Archived Post]

[JRMeyer]

>>> othiele
[December 8, 2020, 2:24pm]

Did you read what's in the link?

[Archived Post]

[JRMeyer]

>>> Toaster1337
[December 8, 2020, 4:37pm]

Now yes,

It running it on Ubuntu 20.04 slash
Python version 3.7 slash
I Think its a running iterference

How I said I use PyAudio to record my mic to a variable (audio) slash
my goal ist go geht the output (Text) to a variable in return.

the Errorcodes you can see in my first Post

I hope thats besser to understand

[Archived Post]

[JRMeyer]

>>> othiele
[December 8, 2020, 4:44pm]

Why don't you do like in the Python
example?

[Archived Post]

[JRMeyer]

>>> Toaster1337
[December 8, 2020, 7:54pm]

Okay not the error is away but deepspeech return always None.

My code:

recorder.py slash
class Recorder: slash
def init(self): slash
slash _pyAudio = pyaudio.PyAudio() slash
self. slash _stream = slash _pyAudio.open(rate=RATE, slash
channels=CHANNELS, slash
format=FORMAT, slash
input=True, slash
output=True, slash
input_device_index=None, slash
output_device_index=0, slash
frames_per_buffer=CHUNK)

self.stt = speechtotextengines.DeepSpeech('./models/deepspeech-0.9.2-models.tflite',
'./models/deepspeech-0.9.2-models.scorer')

staticmethod
def rms(frame):
shorts = array.array(FRAME_FORMAT, frame)
sum_squares = sum(
(sample DEEPSPEECH.cdx deepspeech.commands DEEPSPEECH.pages DEEPSPEECH.warc.gz discourse.mozilla.org html-to-markdown.sh shell-conver-html-to-split-posts.sh sorted-deepspeech-posts NORMALIZE) DEEPSPEECH.cdx deepspeech.commands DEEPSPEECH.pages DEEPSPEECH.warc.gz discourse.mozilla.org html-to-markdown.sh shell-conver-html-to-split-posts.sh sorted-deepspeech-posts 2
for sample in shorts
)
return (sum_squares / len(shorts)) DEEPSPEECH.cdx deepspeech.commands DEEPSPEECH.pages DEEPSPEECH.warc.gz discourse.mozilla.org html-to-markdown.sh shell-conver-html-to-split-posts.sh sorted-deepspeech-posts 0.5 DEEPSPEECH.cdx deepspeech.commands DEEPSPEECH.pages DEEPSPEECH.warc.gz discourse.mozilla.org html-to-markdown.sh shell-conver-html-to-split-posts.sh sorted-deepspeech-posts 1000

def record(self):
print('Noise detected')
rec = []
current = time.time()
end = time.time() + TIMEOUT_LENGTH

while current <= end:

data = self._stream.read(CHUNK)
if self.rms(data) >= THRESHOLD:
end = time.time() + TIMEOUT_LENGTH

current = time.time()
rec.append(data)

# self.stt.run(b''.join(rec))
'''Play Stream'''
# self._stream.write(b''.join(rec))
text = self.stt.run(b''.join(rec))
print(text)
print('Listening...')

def listen(self):
print('Listening...')
while True:
mic_input = self._stream.read(CHUNK)
rms_val = self.rms(mic_input)
if rms_val >= THRESHOLD:
self.record()

if name == 'main':
recorder = Recorder()
recorder.listen()

speechtotextengines.py slash
import numpy slash
import deepspeech slash
import wave

class DeepSpeech:
'''Class to perform speech-to-text transcription and related functionality'''

def init(self, model_path, scorer_path):
self.model = deepspeech.Model(model_path)
self.model.enableExternalScorer(scorer_path)

def run(self, audio):
print('start')
self.model.stt(numpy.frombuffer(audio, numpy.int16))
print('end')

[Archived Post]