def cronista_transcribe(audio_source_path: str,
destination_folder: str,
block_of_transcription: int,
lang: str,
on_transcription_progress=None,
to_file: bool = True):
audio_segments = split(audio_source_path, destination_folder,
block_of_transcription, to_file)
transcribe(audio_segments, lang, on_transcription_progress)
def browse(self):
'''browse for file'''
if self.wordlabel != "": # clear the labels if necessary
self.wordlabel.config(text="")
self.pathlabel.config(text="")
self.timestamplabel.config(text="")
self.estimatelabel.config(text="")
self.filename = askopenfilename(
) # openfile dialog and put file in filename
if not self.filename: # leave method if cancel is clicked
return
self.pathlabel.config(text=basename(
self.filename)) # show filename as label
if self.filename.endswith('.txt'):
return
self.workinglabel.config(text="WORKING",
font=("Helvetica",
20)) # Show WORKING when transcribing
self.pbar_det.pack() # show the progress bar
self.pbar_det.start() # Start the progress bar
root.update()
self.transcription = transcribe(self.filename) # Call transcribe
wordcloud_path = wordcloud_create(self.transcription)
self.transcription = removerlap(self.transcription.split(' '))
self.new_image(wordcloud_path)
self.workinglabel.config(text="") # remove working label
self.pbar_det.stop() # Stop progress bar
self.pbar_det.pack_forget() # Remove progress bar
def transcribe_into_paragraphs(wavfile_path, model_dir, minute_increment,
log_stream):
with open(log_stream, 'a') as f:
f.write('\nBeginning transcription\n')
logging.info('Beginning transcription: ' + wavfile_path)
sentences = transcriber.transcribe(wavfile_path, model_dir, log_stream)
increment = minute_increment * 60
paragraphs = []
paragraph = ''
for (timestamp, sentence) in sentences:
paragraph += (' ' + sentence)
if timestamp / increment > 1:
paragraphs.append(paragraph)
increment += increment
paragraph = ''
return paragraphs
def transcription_to_str(file, time_slot):
transcription = transcribe(file_name=file)
str_transcription = ''.join(transcription)
d = '"{}", {}'.format(str_transcription, time_slot)
return d
for line in infile:
line = line.rstrip()
line=line.replace("<sil>", "");
wordSet.add(line)
wordSet.remove("")
wordList = sorted(list(wordSet))
wordsStr = "\n".join(wordList)
print ">>>> wordsStr\n" + wordsStr
#automagical transformation
transcriber = transcriber.Transcriber()
phonemeSet = set([])
with open("../target/liepa.dic", "wb") as outfile:
for i in range(len(wordList)):
key = wordList[i]
value = transcriber.transcribe(key)
phonemeSet.update(value.split(" "))
#print key +"\t"+ value
outfile.write(key +"\t"+ value + "\n")
with open("../target/liepa.phone", "wb") as outfile:
outfile.write("SIL\n")
for phone in sorted(phonemeSet):
outfile.write(phone + "\n")
def transcription_to_str(file, time_slot):
transcription = transcribe(file_name=file)
d = {}
d['time_slot'] = time_slot
d['str'] = ''.join(transcription)
return d
play("beep_lo.wav")
if 'print' in args.ui_out:
print 'Stopped recording, saving as file #{0} ...'.format(counter)
file_out = os.path.join(dir_out, str(counter) + '.wav')
save_as_wave(file_out, frames, args.channels, args.rate)
counter += 1
except EOFError:
destroy()
print ''
except KeyboardInterrupt:
destroy()
print ''
data = get_all_data(dir_out, counter)
if data:
file_out = os.path.join(dir_out, 'all.wav')
save_data_as_wave(file_out, data, counter)
if args.transcribe:
# TODO: make more configurable
file_out_transcription = os.path.join(dir_out, 'all.txt')
transcription = transcriber.transcribe(file_out, args.uri_transcriber, '', 32000, None, None)
with open(file_out_transcription, 'w') as text_file:
#print(transcription, file=text_file)
text_file.write(transcription)
if args.playback:
play(file_out)
if args.nokeep:
shutil.rmtree(dir_out)