In [1]:
import IPython.display
IPython.display.YouTubeVideo('yZV-XCSmWQY')
Out[1]:
From the same API console, choose "Dashboard" on the left-hand menu and "Enable API".
Enable the following APIs for your project (search for them) if they are not already enabled:
Finally, because we are calling the APIs from Python (clients in many other languages are available), let's install the Python package (it's not installed by default on Datalab)
In [1]:
# Copyright 2016 Google Inc. Licensed under the Apache License, Version 2.0 (the "License");
# !pip install --upgrade google-api-python-client
In [3]:
!pip install --upgrade google-cloud-speech
In [4]:
# Install library into computer storage:
# Below is for cloud AI GCP Datalab VM only: install audio conversion tool
# retcode = subprocess.call(['apt-get', 'update', '-y'])
# retcode = subprocess.call(['apt-get', 'install', 'libav-tools', '-y'])
In [5]:
# Load library into computer memory:
import io
import os
# Imports the Google Cloud client library
from google.cloud import speech
from google.cloud.speech import enums
from google.cloud.speech import types
In [6]:
# Speech-to-Text API
# (1) Instantiates a client - using GOOGLE_APPLICATION_CREDENTIALS
# client_speech2text = speech.SpeechClient()
# (2) Instantiates a client - using 'service account json' file
client_speech2text = speech.SpeechClient.from_service_account_json(
"/media/sf_vm_shared_folder/000-cloud-api-key/mtech-ai-7b7e049cf5f6.json")
In [7]:
# Flag to indicate the environment to run this program:
# Uncomment to run the code on Google Cloud Platform
# parm_runtime_env_GCP = True
# Uncomment to run the code in local machine
parm_runtime_env_GCP = False
import subprocess
# Utility function to convert mp3 file to target GCP audio file type:
# audio_type = ['flac', 'wav']
# audio_file_input = msg['FileName']
# Running Speech API
def didi_mp3_audio_conversion(audio_file_input, audio_type='flac'):
audio_file_output = str(audio_file_input) + '.' + str(audio_type)
# convert mp3 file to target GCP audio file:
# remove audio_file_output, if exists
retcode = subprocess.call(['rm', audio_file_output])
if parm_runtime_env_GCP: # using Datalab in Google Cloud Platform
# GCP: use avconv to convert audio
retcode = subprocess.call(['avconv', '-i', audio_file_input, '-ac', '1', audio_file_output])
else: # using an iss-vm Virtual Machine, or local machine
# VM : use ffmpeg to convert audio
retcode = subprocess.call(['ffmpeg', '-i', audio_file_input, '-ac', '1', audio_file_output])
if retcode == 0:
print('[ O K ] Converted audio file for API: %s' % audio_file_output)
else:
print('[ ERROR ] Function: didi_mp3_audio_conversion() Return Code is : {}'.format(retcode))
return audio_file_output # return file name string only
https://cloud.google.com/speech/docs/languages
'af-ZA' 'Afrikaans (South Africa)' 'id-ID' 'Indonesian (Indonesia)' 'ms-MY' 'Malay (Malaysia)' 'ca-ES' 'Catalan (Spain)' 'cs-CZ' 'Czech (Czech Republic)' 'da-DK' 'Danish (Denmark)' 'de-DE' 'German (Germany)' 'en-AU' 'English (Australia)' 'en-CA' 'English (Canada)' 'en-GB' 'English (United Kingdom)' 'en-IN' 'English (India)' 'en-IE' 'English (Ireland)' 'en-NZ' 'English (New Zealand)' 'en-PH' 'English (Philippines)' 'en-ZA' 'English (South Africa)' 'en-US' 'English (United States)' 'es-AR' 'Spanish (Argentina)' 'es-BO' 'Spanish (Bolivia)' 'es-CL' 'Spanish (Chile)' 'es-CO' 'Spanish (Colombia)' 'es-CR' 'Spanish (Costa Rica)' 'es-EC' 'Spanish (Ecuador)' 'es-SV' 'Spanish (El Salvador)' 'es-ES' 'Spanish (Spain)' 'es-US' 'Spanish (United States)' 'es-GT' 'Spanish (Guatemala)' 'es-HN' 'Spanish (Honduras)' 'es-MX' 'Spanish (Mexico)' 'es-NI' 'Spanish (Nicaragua)' 'es-PA' 'Spanish (Panama)' 'es-PY' 'Spanish (Paraguay)' 'es-PE' 'Spanish (Peru)' 'es-PR' 'Spanish (Puerto Rico)' 'es-DO' 'Spanish (Dominican Republic)' 'es-UY' 'Spanish (Uruguay)' 'es-VE' 'Spanish (Venezuela)' 'eu-ES' 'Basque (Spain)' 'fil-PH' 'Filipino (Philippines)' 'fr-CA' 'French (Canada)' 'fr-FR' 'French (France)' 'gl-ES' 'Galician (Spain)' 'hr-HR' 'Croatian (Croatia)' 'zu-ZA' 'Zulu (South Africa)' 'is-IS' 'Icelandic (Iceland)' 'it-IT' 'Italian (Italy)' 'lt-LT' 'Lithuanian (Lithuania)' 'hu-HU' 'Hungarian (Hungary)' 'nl-NL' 'Dutch (Netherlands)' 'nb-NO' 'Norwegian Bokmål (Norway)' 'pl-PL' 'Polish (Poland)' 'pt-BR' 'Portuguese (Brazil)' 'pt-PT' 'Portuguese (Portugal)' 'ro-RO' 'Romanian (Romania)' 'sk-SK' 'Slovak (Slovakia)' 'sl-SI' 'Slovenian (Slovenia)' 'fi-FI' 'Finnish (Finland)' 'sv-SE' 'Swedish (Sweden)' 'vi-VN' 'Vietnamese (Vietnam)' 'tr-TR' 'Turkish (Turkey)' 'el-GR' 'Greek (Greece)' 'bg-BG' 'Bulgarian (Bulgaria)' 'ru-RU' 'Russian (Russia)' 'sr-RS' 'Serbian (Serbia)' 'uk-UA' 'Ukrainian (Ukraine)' 'he-IL' 'Hebrew (Israel)' 'ar-IL' 'Arabic (Israel)' 'ar-JO' 'Arabic (Jordan)' 'ar-AE' 'Arabic (United Arab Emirates)' 'ar-BH' 'Arabic (Bahrain)' 'ar-DZ' 'Arabic (Algeria)' 'ar-SA' 'Arabic (Saudi Arabia)' 'ar-IQ' 'Arabic (Iraq)' 'ar-KW' 'Arabic (Kuwait)' 'ar-MA' 'Arabic (Morocco)' 'ar-TN' 'Arabic (Tunisia)' 'ar-OM' 'Arabic (Oman)' 'ar-PS' 'Arabic (State of Palestine)' 'ar-QA' 'Arabic (Qatar)' 'ar-LB' 'Arabic (Lebanon)' 'ar-EG' 'Arabic (Egypt)' 'fa-IR' 'Persian (Iran)' 'hi-IN' 'Hindi (India)' 'th-TH' 'Thai (Thailand)' 'ko-KR' 'Korean (South Korea)' 'cmn-Hant-TW' 'Chinese, Mandarin (Traditional, Taiwan)' 'yue-Hant-HK' 'Chinese, Cantonese (Traditional, Hong Kong)' 'ja-JP' 'Japanese (Japan)' 'cmn-Hans-HK' 'Chinese, Mandarin (Simplified, Hong Kong)' 'cmn-Hans-CN' 'Chinese, Mandarin (Simplified, China)'
In [8]:
# API control parameter for 语音转换成消息文字 (Speech recognition: voice to text)
parm_speech_recognition_language = 'en-US' # speech API 'voice to text' language
# parm_speech_recognition_language = 'cmn-Hans-CN' # speech API 'voice to text' language
In [9]:
didi_mp3_audio_conversion('reference/S-IPA-welcome.mp3', 'wav')
Out[9]:
In [10]:
didi_mp3_audio_conversion('reference/S-IPA-welcome.mp3')
Out[10]:
In [11]:
didi_mp3_audio_conversion('reference/S-IPA-welcome.m4a')
Out[11]:
In [12]:
def didi_speech2text(speech_file, didi_language_code='en-US'):
"""Transcribe the given audio file."""
from google.cloud import speech
from google.cloud.speech import enums
from google.cloud.speech import types
##################################################################
# client_speech2text = speech.SpeechClient()
#
# client_speech2text = speech.SpeechClient.from_service_account_json(
# "/media/sf_vm_shared_folder/000-cloud-api-key/mtech-ai-7b7e049cf5f6.json")
##################################################################
with io.open(speech_file, 'rb') as audio_file:
content = audio_file.read()
audio = types.RecognitionAudio(content=content)
# print('audio : ', audio)
config = types.RecognitionConfig(
# encoding=enums.RecognitionConfig.AudioEncoding.FLAC,
# encoding=enums.RecognitionConfig.AudioEncoding.LINEAR16,
# sample_rate_hertz=16000,
language_code=didi_language_code)
response = client_speech2text.recognize(config, audio)
print('response : ', response)
# Each result is for a consecutive portion of the audio. Iterate through
# them to get the transcripts for the entire audio file.
for result in response.results:
# The first alternative is the most likely one for this portion.
print(u'Transcript : {}'.format(result.alternatives[0].transcript))
print(u'Confidence : {0:.3f}'.format(result.alternatives[0].confidence))
return response
In [13]:
response = didi_speech2text(didi_mp3_audio_conversion('reference/S-IPA-welcome.mp3'), parm_speech_recognition_language)
In [14]:
response = didi_speech2text(didi_mp3_audio_conversion('reference/S-IPA-welcome.mp3', 'wav'), parm_speech_recognition_language)
In [15]:
response = didi_speech2text(didi_mp3_audio_conversion('reference/S-IPA-welcome.m4a'), parm_speech_recognition_language)
In [16]:
response.results
Out[16]:
In [17]:
response.results[0]
Out[17]:
In [18]:
response.results[0].alternatives
Out[18]:
In [19]:
response.results[0].alternatives[0]
Out[19]:
In [20]:
response.results[0].alternatives[0].transcript
Out[20]:
In [21]:
response.results[0].alternatives[0].confidence
Out[21]:
This page demonstrates how to transcribe long audio files (longer than 1 minute) to text using asynchronous speech recognition.
https://cloud.google.com/speech-to-text/docs/async-recognize
In [22]:
!pip install --upgrade google-cloud-texttospeech
In [23]:
# Load library into computer memory:
import io
import os
# Imports the Google Cloud client library
from google.cloud import texttospeech
In [24]:
# 2018 Oct
# Cuttently Google Text-to-Speech API texttospeech.TextToSpeechClient can only use GOOGLE_APPLICATION_CREDENTIALS
!echo $GOOGLE_APPLICATION_CREDENTIALS
In [25]:
# Text-to-Speech API
# (1) Instantiates a client - using GOOGLE_APPLICATION_CREDENTIALS
client_text2speech = texttospeech.TextToSpeechClient()
# (2) Instantiates a client - using 'service account json' file
# client_text2speech = texttospeech.TextToSpeechClient.from_service_account_json(
# "/media/sf_vm_shared_folder/000-cloud-api-key/mtech-ai-7b7e049cf5f6.json")
https://cloud.google.com/text-to-speech/docs/voices
Dutch (Netherlands), English (Australia), English (UK), English (US), French, French (Canada), German, Italian, Japanese, Korean, Portugese (Brazil), Spanish, Swedish, Turkish
In [26]:
def didi_text2speech(input_text="Hello, World!", output_mp3_file_name='text2speech.mp3', didi_language_code='en-US'):
"""Synthesizes speech from the input string of text or ssml.
Note: ssml must be well-formed according to:
https://www.w3.org/TR/speech-synthesis/
"""
# Set the text input to be synthesized
synthesis_input = texttospeech.types.SynthesisInput(text=input_text)
# Build the voice request, select the language code ("en-US") and the ssml
# voice gender ("neutral")
voice = texttospeech.types.VoiceSelectionParams(
language_code=didi_language_code,
ssml_gender=texttospeech.enums.SsmlVoiceGender.NEUTRAL)
# Select the type of audio file you want returned
audio_config = texttospeech.types.AudioConfig(
audio_encoding=texttospeech.enums.AudioEncoding.MP3)
# Perform the text-to-speech request on the text input with the selected
# voice parameters and audio file type
response = client_text2speech.synthesize_speech(synthesis_input, voice, audio_config)
# The response's audio_content is binary.
with open(output_mp3_file_name, 'wb') as out:
# Write the response to the output file.
out.write(response.audio_content)
print('Audio content written to mp3 file : {}'.format(output_mp3_file_name))
return output_mp3_file_name
In [27]:
# API control parameter for 消息文字转成语音 (Speech synthesis: text to voice)
parm_speech_synthesis_language = 'en-US' # speech synthesis API 'text to voice' language
# parm_speech_synthesis_language = 'es-ES' # speech synthesis API 'text to voice' language
In [28]:
input_text='Hi everyone! It\'s great to meet you in intelligent process automation course!'
didi_text2speech(input_text, 'reference/text2speech_IPA.mp3', parm_speech_synthesis_language)
Out[28]:
In [29]:
# Local AI Module for Speech Synthesis: Text-to-Speech
# Install library into computer storage:
# Below is an optional local AI module for "text to speech"
# import subprocess
# retcode = subprocess.call(['pip', 'install', '-U', 'gTTS'])
!pip install -U gTTS
# Load library into computer memory:
# Below is for Google Speech synthesis: text to voice API
from gtts import gTTS
https://pypi.python.org/pypi/gTTS
The easiest way to get a list of available language is to print them with: gtts-cli --all
iss-user@iss-vm:~$ gtts-cli --all
af: Afrikaans ar: Arabic bn: Bengali bs: Bosnian ca: Catalan cs: Czech cy: Welsh da: Danish de: German el: Greek en-au: English (Australia) en-ca: English (Canada) en-gb: English (UK) en-gh: English (Ghana) en-ie: English (Ireland) en-in: English (India) en-ng: English (Nigeria) en-nz: English (New Zealand) en-ph: English (Philippines) en-tz: English (Tanzania) en-uk: English (UK) en-us: English (US) en-za: English (South Africa) en: English eo: Esperanto es-es: Spanish (Spain) es-us: Spanish (United States) es: Spanish et: Estonian fi: Finnish fr-ca: French (Canada) fr-fr: French (France) fr: French hi: Hindi hr: Croatian hu: Hungarian hy: Armenian id: Indonesian is: Icelandic it: Italian ja: Japanese jw: Javanese km: Khmer ko: Korean la: Latin lv: Latvian mk: Macedonian ml: Malayalam mr: Marathi my: Myanmar (Burmese) ne: Nepali nl: Dutch no: Norwegian pl: Polish pt-br: Portuguese (Brazil) pt-pt: Portuguese (Portugal) pt: Portuguese ro: Romanian ru: Russian si: Sinhala sk: Slovak sq: Albanian sr: Serbian su: Sundanese sv: Swedish sw: Swahili ta: Tamil te: Telugu th: Thai tl: Filipino tr: Turkish uk: Ukrainian vi: Vietnamese zh-cn: Chinese (Mandarin/China) zh-tw: Chinese (Mandarin/Taiwan)
In [30]:
# API control parameter for 消息文字转成语音 (Speech synthesis: text to voice)
parm_speech_synthesis_language_local = 'en' # speech synthesis API 'text to voice' language
In [31]:
import sys
# Running Local AI Module Text-to-Speech
def didi_text2speech_local(input_text="Hello, World!", output_mp3_file_name='text2speech_local.mp3', didi_language_code='en'):
# Python 2
if sys.version_info[0] < 3:
tts = gTTS(text=input_text.encode('utf-8'), lang=didi_language_code, slow=False)
# Python 3
else:
tts = gTTS(text=input_text, lang=didi_language_code, slow=False)
tts.save(output_mp3_file_name)
print('\nCompeleted: Speech synthesis API ( Text -> Voice)')
print(input_text)
return output_mp3_file_name
In [32]:
input_text='Hi everyone! It\'s great to meet you in intelligent process automation course!'
didi_text2speech_local(input_text, 'reference/text2speech_IPA_local.mp3', parm_speech_synthesis_language_local)
Out[32]:
In [33]:
# API control parameter for 消息文字转成语音 (Speech synthesis: text to voice)
parm_speech_synthesis_language_local = 'zh-cn' # speech synthesis API 'text to voice' language
In [34]:
input_text="明月几时有？把酒问青天。不知天上宫阙，今夕是何年？我欲乘风归去，又恐琼楼玉宇，高处不胜寒。起舞弄清影，何似在人间？转朱阁，低绮户，照无眠。不应有恨，何事长向别时圆？人有悲欢离合，月有阴晴圆缺，此事古难全。但愿人长久，千里共婵娟。"
didi_text2speech_local(input_text, 'reference/text2speech_local_zh.mp3', parm_speech_synthesis_language_local)
Out[34]:
In [35]:
# API control parameter for 语音转换成消息文字 (Speech recognition: voice to text)
parm_speech_recognition_language = 'en-US' # speech API 'voice to text' language
# parm_speech_recognition_language = 'cmn-Hans-CN' # speech API 'voice to text' language
# API control parameter for 消息文字转成语音 (Speech synthesis: text to voice)
########### Cloud AI API ###########
parm_speech_synthesis_language = 'en-US' # speech synthesis API 'text to voice' language
# parm_speech_synthesis_language = 'es-ES' # speech synthesis API 'text to voice' language
########### Local AI Module gTTS ###########
parm_speech_synthesis_language_local = 'en' # speech synthesis API 'text to voice' language
# parm_speech_synthesis_language_local = 'zh-cn' # speech synthesis API 'text to voice' language
In [36]:
import re
import itchat
from itchat.content import *
In [40]:
# itchat.auto_login(hotReload=True) # hotReload=True: 退出程序后暂存登陆状态。即使程序关闭，一定时间内重新开启也可以不用重新扫码。
itchat.auto_login(enableCmdQR=-2) # enableCmdQR=-2: 命令行显示QR图片
In [41]:
# Obtain my own Nick Name
MySelf = itchat.search_friends()
NickName4RegEx = '@' + MySelf['NickName'] + '\s*'
In [43]:
# 1. 语音转换成消息文字 (Speech recognition: voice to text)
@itchat.msg_register([RECORDING], isGroupChat=True)
@itchat.msg_register([RECORDING])
def download_files(msg):
msg.download(msg.fileName)
print('\nDownloaded audio file name is: %s' % msg['FileName'])
##############################################################################################################
# call audio analysis APIs #
##############################################################################################################
audio_analysis_reply = u'[ Audio Analysis 音频处理结果 ]\n'
# Voice to Text:
audio_analysis_reply += u'\n[ Voice -> Text 语音识别 ]\n'
# response = KudosData_voice_to_text(msg['FileName'], 'flac')
response = didi_speech2text(didi_mp3_audio_conversion(msg['FileName']), parm_speech_recognition_language)
audio_analysis_reply += str(response) + u'\n'
return audio_analysis_reply
In [44]:
# 2.1 消息文字转成语音 (Speech synthesis: text to voice)
########### Cloud AI API ###########
# 在群里，如果收到 @ 自己的文字信息，会自动将文字转换成语音，再以 mp3 文件方式发送回复：
@itchat.msg_register(TEXT, isGroupChat=True)
def text_to_voice_reply_1(msg):
if msg['isAt']:
# Remove my own Nick Name from message:
text2voice = re.sub(NickName4RegEx, '', msg['Content'])
text2voiceMP3name = KudosData_text_to_voice(text2voice)
text2voiceMP3name = didi_text2speech(text2voice, 'va_t2s.mp3', parm_speech_synthesis_language)
itchat.send('@%s@%s' % ('fil', text2voiceMP3name), msg['FromUserName'])
In [45]:
# 2.2 消息文字转成语音 (Speech synthesis: text to voice)
########### Local AI Module gTTS ###########
# 在群里，如果收到 @ 自己的文字信息，会自动将文字转换成语音，再以 mp3 文件方式发送回复：
@itchat.msg_register(TEXT, isGroupChat=False)
def text_to_voice_reply_2(msg):
# Remove my own Nick Name from message:
text2voice = re.sub(NickName4RegEx, '', msg['Content'])
text2voiceMP3name = didi_text2speech_local(text2voice, 'va_t2s_local.mp3', parm_speech_synthesis_language_local)
itchat.send('@%s@%s' % ('fil', text2voiceMP3name), msg['FromUserName'])
In [46]:
itchat.run()
In [47]:
# interupt kernel, then logout
itchat.logout() # 安全退出
Out[47]: