GoogleColab

vosk_test_googletrans,py

#!/usr/bin/env python3

from vosk import Model, KaldiRecognizer, SetLogLevel
import sys
import os
import wave
import json
from googletrans import Translator
import h2.connection
import h2.config
import datetime


path = '/content/vosk-api/python/example/'

SetLogLevel(-1)
#SetLogLevel(0)

if not os.path.exists("model"):
    print ("Please download the model from https://alphacephei.com/vosk/models and unpack as 'model' in the current folder.")
    exit (1)

#wf = wave.open(path+'/test.wav',"rb")#English test sample
wf = wave.open(path+'/test1.wav',"rb")#Chinese lang test sample
sound = path+'/test1.wav'
if wf.getnchannels() != 1 or wf.getsampwidth() != 2 or wf.getcomptype() != "NONE":
    print ("Audio file must be WAV format mono PCM.")
    exit (1)

model = Model("model")
rec = KaldiRecognizer(model, wf.getframerate())

#obj_num = 1

def composit(line):
    ##2 lines to 1 line 
    extract = line

    text_compo = []
    txt = ''
    for count,text in enumerate(extract):
      if (count+1)%2 == 0:
        txt = text_compo.pop()
        txt += ' ' +text
        text_compo.append(txt)
      else :
        txt = text
        text_compo.append(txt)
    return text_compo
    #print(text_compo)  

def google_a(line):
    line[:] = [l.strip() for l in line]
    line[:] = [l.rstrip('\n') for l in line]
    line[:] = [l.rstrip('\r') for l in line]
    line[:] = [a for a in line if a != '']
    line[:] = [l.replace('\n',' ') for l in line]
    
    #translator = Translator()
    num = 5
#    global obj_num
    print()
    
    print('contents')
    print()

    for count, l in enumerate(line):
        if count +1< num:
            ln = str(count+1)
            ln = ln.zfill(4)
            #print('cn ',ln,'  ', l) # original text
            print('en ',ln,'  ', l) # original text
        else:
            num += 5
            ln = str(count+1)
            ln = ln.zfill(4)
            #print('cn ',ln,'  ',l) # original text
            print('en ',ln,'  ', l) # original text
   
#            obj_num = obj_num + 1
#            print()
#            print("--- translator :", obj_num)
#            print()
    #del translator
    
    num = 5
    print()
    translator = Translator()
#    obj_num += 1
    for count, l in enumerate(line):
        if count +1< num:
            translated = translator.translate(l, dest='ja')
            #print('cn ',count+1,'  ', l) # original text
            ln = str(count+1)
            ln = ln.zfill(4)
            print('ja ',ln,'  ',translated.text)
        else:
            num += 5
            translated = translator.translate(l, dest='ja')
            #print('cn ',count+1,'  ',l) # original text
            ln = str(count+1)
            ln = ln.zfill(4)
            print('ja ',ln,'  ',translated.text)        
            del translator
   
#            obj_num = obj_num + 1
#            print()
#            print("--- translator :", obj_num)
#            print()
            translator = Translator()
    del translator
    
    print()
    translator = Translator()
    num = 5
#    obj_num += 1
    for count, l in enumerate(line):
        if count +1< num:
            translated = translator.translate(l, dest='zh-tw')
            #print('cn ',count+1,'  ', l) # original text
            ln = str(count+1)
            ln = ln.zfill(4)
            print('tw ',ln,'  ',translated.text)
        else:
            num += 5
            translated = translator.translate(l, dest='zh-tw')
            #print('cn ',count+1,'  ',l) # original text
            ln = str(count+1)
            ln = ln.zfill(4)
            print('tw ',ln,'  ',translated.text)        
            del translator
   
#            obj_num = obj_num + 1
#            print()
#            print("--- translator :", obj_num)
#            print()
            translator = Translator()
    del translator

def google(line):
    line[:] = [l.strip() for l in line]
    line[:] = [l.rstrip('\n') for l in line]
    line[:] = [l.rstrip('\r') for l in line]
    line[:] = [a for a in line if a != '']
    line[:] = [l.replace('\n',' ') for l in line]
    translator = Translator()
    num = 10
#    global obj_num
    
    #print(len(line))
    index = int(len(line))

    for count, l in enumerate(line):
        if count +1< num:
            if count == index - 1:
                translated_1 = translator.translate(l, dest='ja')
                translated_2 = translator.translate(l, dest='zh-tw')
                #translated_3 = translator.translate(l, dest='ko')
                #translated_4 = translator.translate(l, dest='tr')
                #ln = str(count+1)
                #ln = ln.zfill(4)
                #print('cn ',ln,' ', l) # original text
                print('en ',' ', l) # original text
                print('ja ',' ', translated_1.text)
                print('tw ',' ', translated_2.text)
                #print('ko ',' ', translated_3.text)
                #print('tr ',ln,' ', translated_4.text)
            else:
                continue
        else:
            num += 10
            if count == index - 1:
                translated_1 = translator.translate(l, dest='ja')
                translated_2 = translator.translate(l, dest='zh-tw')
                #translated_3 = translator.translate(l, dest='ko')
                #translated_4 = translator.translate(l, dest='tr')
                #ln = str(count+1)
                #ln = ln.zfill(4)
                #print('cn ',ln,' ', l) # original text
                print('en ',' ', l) # original text
                print('ja ',' ',translated_1.text)        
                print('tw ',' ', translated_2.text)
                #print('ko ',' ', translated_3.text)
                #print('tr ',ln,' ', translated_4.text)
                del translator
                
#                obj_num = obj_num + 1
#                print()
#                print("--- translator :", obj_num)
#                print()
                translator = Translator()
            else:
                continue
    del translator

def fmttime(seconds):
    secs = seconds #millisecs / 1000.0
    d = datetime.timedelta(seconds=secs)
    t = (datetime.datetime.min + d).time()
    milli = t.strftime('%f')[:3]
    value = t.strftime('%H:%M:%S,') + milli
    return value

lines = []
line = []
oder = 1
while True:
    data = wf.readframes(4000)
    if len(data) == 0:
        break
    if rec.AcceptWaveform(data):
        #print(rec.Result())
        res = json.loads(rec.Result())
        for tr in res.keys():
            #if isinstance(res[tr], dict) == False:
            #print(tr)
            if tr == 'result':
                #print(rec.Result())
                #print(res['result'][0]['start'])
                starttime = fmttime(res['result'][0]['start'])
                #print(res['result'][-1]['end'])
                endtime = fmttime(res['result'][-1]['end'])
                print(oder)
                oder += 1
                print(starttime,'-->',endtime)
                del starttime,endtime
                if res['text'] != '':
                    line.append(res['text']) 
                    print(*line)
                    ##google(line)
                    lines.append(line.pop())
                    print()
        text_compo = composit(line)
    else:
        #print(rec.PartialResult())
        continue

#print(rec.FinalResult())
res = json.loads(rec.FinalResult())
#print(res['text'])
line.append(res['text'])
print(oder)
starttime = fmttime(res['result'][0]['start'])
endtime = fmttime(res['result'][-1]['end'])
print(starttime,'-->',endtime)
print(*line)
##google(line)
lines.append(line.pop())

#text_compo = composit(lines)

#google_a(text_compo)