bitirme-proje / tryVosk.py
Bedirhan's picture
Upload tryVosk.py
3e70ffd
raw
history blame
3.1 kB
from vosk import Model, KaldiRecognizer
import wave
import json
import os
class vosk():
def __init__(self, file_folder, outputResult, outputText, lang):
self.file_folder = file_folder
self.outputResult = outputResult
self.outputText = outputText
self.lang = lang
def control(self):
if os.path.exists(self.outputResult):
os.remove(self.outputResult)
if os.path.exists(self.outputText):
os.remove(self.outputText)
def useVosk(self, file_name):
wf = wave.open(file_name, "rb")
# initialize a str to hold results
results = ""
textResults = ""
# build the model and recognizer objects.
if lang == "Turkish" or lang == "Türkçe" or lang == "türkçe" or lang == "turkish":
model = Model("vosk-modelss/vosk-model-tr")
elif lang == "English" or lang == "İngilizce" or lang == "ingilizce" or lang == "english":
model = Model("vosk-modelss/vosk-model-en")
else:
print("Dil bulunamadı!------The language has not been found!")
recognizer = KaldiRecognizer(model, wf.getframerate())
recognizer.SetWords(True)
while True:
data = wf.readframes(4000)
if len(data) == 0:
break
if recognizer.AcceptWaveform(data):
recognizerResult = recognizer.Result()
results = results + recognizerResult
results_y = (json.dumps(results, ensure_ascii=False).encode('utf-8')).decode()
resultDict = json.loads(recognizerResult)
# process "final" result
results = results + recognizer.FinalResult()
results_y = (json.dumps(results, ensure_ascii=False).encode('utf-8')).decode()
for i in range(results_y.count('text')):
txtt = results_y.split("text")[i + 1].split("\\n")[0].split('"')[2].split("\\")[0]
textResults += txtt
textResults += " "
resultDict = json.loads(recognizer.FinalResult())
if len(textResults) == 0:
textResults += txtt
return results, textResults
def playWav(self):
self.control()
outputResultt = open(self.outputResult, 'w', encoding="utf-8")
outputTextt = open(self.outputText, 'w', encoding="utf-8")
for i in os.listdir(self.file_folder):
if i.endswith("wav"):
results, textResults = self.useVosk(self.file_folder + i)
# write results to a file
outputResultt.write("%s\n" % results)
# write text portion of results to a file
outputTextt.write("%s\n" % textResults)
print("Metin alımı tamamlandı!----------Text retrieval has been completed!")
outputTextt.close()
outputResultt.close()
file_fold, outputR, outputT = "Sounds/tr/", "texts/resulttr.json", "texts/texttr.txt"
lang = "turkish"
vosk(file_fold, outputR, outputT, lang).playWav()