Digital Nomad

$ mv ~/Downloads/api-project-249965614499-33c4f9f41469.json .

pip install google-api-python-client google-auth-httplib2 google-auth-oauthlib
pip install oauth2client
pip install google-cloud
pip install argparse

from googleapiclient import discovery
from google.oauth2 import service_account
import json
import os

DOCUMENT_ID='1FCGcY00_8_DK8edLP1ws5NQamdmB6agCZpfXXXXXXXXXXXXX'

SCOPES = [
#    'https://www.googleapis.com/auth/drive',
    'https://www.googleapis.com/auth/documents.readonly'
]
# DISCOVERY_DOC = 'https://docs.googleapis.com/$discovery/rest?version=v1'
DISCOVERY_DOC = 'https://docs.googleapis.com/$discovery/rest?version=v1'

SERVICE_ACCOUNT_FILE = 'api-project-249965614499-33c4f9f41469.json'
credentials = service_account.Credentials.from_service_account_file(
        SERVICE_ACCOUNT_FILE, scopes=SCOPES)

service = discovery.build('docs', 'v1', discoveryServiceUrl=DISCOVERY_DOC, credentials=credentials)

# Do a document "get" request and print the results as formatted JSON
f = open("dic.json", "w")
result = service.documents().get(documentId=DOCUMENT_ID).execute()
jsonStr = json.dumps(result, indent=4, sort_keys=True)
f.write(jsonStr)
f.close()

print("dic.json has been created")

{
    "body": {
        "content": [
            {
                "endIndex": 1,
                "sectionBreak": {
                    "sectionStyle": {
                        "columnSeparatorStyle": "NONE",
                        "contentDirection": "LEFT_TO_RIGHT",
                        "sectionType": "CONTINUOUS"
                    }
                }
            },
            {
                "endIndex": 38,
                "paragraph": {
                    "elements": [
                        {
                            "endIndex": 38,
                            "startIndex": 1,
                            "textRun": {
                                "content": "Google Cloud TTS to read Google Docs\n",
                                "textStyle": {}
                            }
                        }
                    ],
                    "paragraphStyle": {
                        "direction": "LEFT_TO_RIGHT",
                        "headingId": "h.swe0qm2wob2",
                        "namedStyleType": "TITLE"
                    }
                },
                "startIndex": 1
            },
...
            {
                "endIndex": 42,
                "paragraph": {
                    "elements": [
                        {
                            "endIndex": 42,
                            "startIndex": 39,
                            "textRun": {
                                "content": "\ucc38\uace0\n",
                                "textStyle": {}
                            }
                        }
                    ],
                    "paragraphStyle": {
                        "direction": "LEFT_TO_RIGHT",
                        "headingId": "h.1o1ly2f6sob",
                        "namedStyleType": "HEADING_2"
                    }
                },
                "startIndex": 39
            },

...

service = discovery.build('docs', 'v1', discoveryServiceUrl=DISCOVERY_DOC, credentials=credentials)

result = service.documents().get(documentId=DOCUMENT_ID).execute()

for sentence in result['body']['content']:
if 'paragraph' in sentence and 'textRun' in sentence['paragraph']['elements'][0]:
print(sentence['endIndex'])
# print("**************************************************************************")
# print(json.dumps(sentence['paragraph'], indent=4, sort_keys=True))
text = sentence['paragraph']['elements'][0]['textRun']['content']
style = sentence['paragraph']['paragraphStyle']['namedStyleType']
print(style, text)

...

for sentence in result['body']['content']:
if 'paragraph' in sentence and 'textRun' in sentence['paragraph']['elements'][0]:
print(sentence['endIndex'])
# print("**************************************************************************")
# print(json.dumps(sentence['paragraph'], indent=4, sort_keys=True))
text = sentence['paragraph']['elements'][0]['textRun']['content']
style = sentence['paragraph']['paragraphStyle']['namedStyleType']
print(style, text)
elif 'table' in sentence:
print(sentence['endIndex'])
for tableRow in sentence['table']['tableRows']:
for tableCell in tableRow['tableCells']:
text = ''
for tableContents in tableCell['content']:
print(tableContents['endIndex'])
for tableSentence in tableContents['paragraph']['elements']:
text = text + tableSentence['textRun']['content']
text = text + '\u000b'
for line in text.split('\u000b'):
print('[TABLE]', line)

컬럼1

컬럼2

컬럼3

로우2-1

로우2-2

로우2-3

로구3-1

로우3-2

로우3-3

문장인 경우에는 어떻게 하는 지 궁금합니다. 이어진 문장.

떨어진 문장.

from flask import Flask, request, Response
from googleapiclient import discovery
from google.oauth2 import service_account
import json
import os

app = Flask(__name__)

@app.route("/<docid>")
def hello(docid):
styles = {
'TITLE': '<h1>%s%s</h1>',
'HEADING_1': '<h2>%s%s</h2>',
'HEADING_2': '<h3>%s%s</h3>',
'HEADING_3': '<h4>%s%s</h4>',
'NORMAL_TEXT': '<p>%s%s</p>'
}

DOCUMENT_ID=docid

SCOPES = [
#    'https://www.googleapis.com/auth/drive',
'https://www.googleapis.com/auth/documents.readonly'
]
# DISCOVERY_DOC = 'https://docs.googleapis.com/$discovery/rest?version=v1'
DISCOVERY_DOC = 'https://docs.googleapis.com/$discovery/rest?version=v1'

SERVICE_ACCOUNT_FILE = 'api-project-249965614499-33c4f9f41469.json'
credentials = service_account.Credentials.from_service_account_file(
SERVICE_ACCOUNT_FILE, scopes=SCOPES)

service = discovery.build('docs', 'v1', discoveryServiceUrl=DISCOVERY_DOC, credentials=credentials)

result = service.documents().get(documentId=DOCUMENT_ID).execute()

f = open("dic.json", "w")
jsonStr = json.dumps(result, indent=4, sort_keys=True)
f.write(jsonStr)
f.close()

html = ''
for sentence in result['body']['content']:
if 'paragraph' in sentence and 'textRun' in sentence['paragraph']['elements'][0]:
print(sentence['endIndex'])
# print("**************************************************************************")
# print(json.dumps(sentence['paragraph'], indent=4, sort_keys=True))
bullet = ''
if 'bullet' in sentence['paragraph']:
bullet = '&#8226;'
text = ''
for t in sentence['paragraph']['elements']:
text = text + t['textRun']['content']
style = sentence['paragraph']['paragraphStyle']['namedStyleType']
print(style, text)
html = html + (styles[style] % (bullet, text)) + "\n"
elif 'table' in sentence:
print(sentence['endIndex'])
html = html + "<table border='1'>\n"
for tableRow in sentence['table']['tableRows']:
html = html + "<tr>\n"
for tableCell in tableRow['tableCells']:
html = html + "<td>\n"
text = ''
for tableContents in tableCell['content']:
print(tableContents['endIndex'])
for tableSentence in tableContents['paragraph']['elements']:
text = text + tableSentence['textRun']['content']
text = text + '\u000b'
for line in text.split('\u000b'):
print('[TABLE]', line)
html = html + ("<p>%s</p>\n" % (line))
html = html + "</td>\n"
html = html + "</tr>\n"
return html, 200

if __name__ == "__main__":
app.run(host='0.0.0.0', port=5001, debug=False)

styles = {
'TITLE': '<div class="jerry_tts"><h1>%s%s</h1></div>',
'HEADING_1': '<div class="jerry_tts"><h2>%s%s</h2></div>',
'HEADING_2': '<div class="jerry_tts"><h3>%s%s</h3></div>',
'HEADING_3': '<div class="jerry_tts"><h4>%s%s</h4></div>',
'NORMAL_TEXT': '<div class="jerry_tts"><p>%s%s</p></div>'
}

$ pip install google-cloud-texttospeech

@app.route("/speak/<encoding>")
def speak(encoding):
text = request.args.get('text', '')

# Instantiates a client
os.environ['GOOGLE_APPLICATION_CREDENTIALS'] = SERVICE_ACCOUNT_FILE
client = texttospeech.TextToSpeechClient()

# Set the text input to be synthesized
synthesis_input = texttospeech.SynthesisInput(text=text)

# Build the voice request, select the language code ("en-US") and the ssml
# voice gender ("neutral")
voice = texttospeech.VoiceSelectionParams(
language_code=encoding, ssml_gender=texttospeech.SsmlVoiceGender.NEUTRAL
)

# Select the type of audio file you want returned
audio_config = texttospeech.AudioConfig(
audio_encoding=texttospeech.AudioEncoding.MP3
)

# Perform the text-to-speech request on the text input with the selected
# voice parameters and audio file type
response = client.synthesize_speech(
input=synthesis_input, voice=voice, audio_config=audio_config
)

# The response's audio_content is binary.
with open("output.mp3", "wb") as out:
# Write the response to the output file.
out.write(response.audio_content)

return Response(response=response.audio_content, mimetype='audio/mp3')

scripts = '''
<script>
idx = -1;
lastIdx = -1;
var audio = null;

window.onload = function() {
document.addEventListener('keydown', keyEvent);
setTimeout(forward, 200);
}

function play() {
sentences = document.querySelectorAll(".jerry_tts");
sentence = sentences[idx];
if(sentence.nodeName == "DIV" && sentence.textContent) {
if(lastIdx >= 0) {
sentences[lastIdx].style.backgroundColor='transparent';
}
sentence.style.backgroundColor='yellow';
lastIdx = idx;
sentence.scrollIntoView(true);
window.scrollTo(0, window.scrollY - 100);
encoding = document.querySelector("body > select:nth-child(1)").value
speed = document.querySelector("body > select:nth-child(2)").value
if(audio) {
audio.pause();
}
audio = new Audio("/speak/" + encoding + "?text=" + sentence.textContent);
audio.load();
audio.playbackRate = speed;
audio.play()
.then(() => {
// Playing
})
.catch(error => {
console.log("empty voice");
setTimeout(forward, 100);
});
audio.addEventListener('ended', forward);
}
}

function forward() {
sentences = document.querySelectorAll(".jerry_tts");
if(idx - 1 < sentences.length) {
idx++;
}

play();
}

function backward() {
if(idx > 0) {
idx--;
}

play();
}

function pause() {
if(audio) {
if(audio.paused) {
audio.play();
}
else {
audio.pause();
}
}
}

function keyEvent(e) {
    e = e || window.event;
if(e.ctrlKey && e.key == '.') {
forward();
}
else if(e.ctrlKey && e.key == ',') {
backward();
}
else if(e.ctrlKey && e.key == '/') {
pause();
}
}
</script>
'''

  html = '<html>\n'
  html = html + scripts
  html = html + '''
      <body>
      <select name='encoidng' onChange='play()'>
          <option value='en-US'>en-US</option>
          <option value='ko-KR'>ko-KR</option>
      </select>
      <select name='speed' onChange='play()'>
          <option value='2'>2배</option>
          <option value='1.5'>1.5배</option>
          <option value='1.3'>1.3배</option>
          <option value='1.2'>1.2배</option>
          <option value='1.1'>1.1배</option>
          <option value='1' selected='selected'>1배</option>
          <option value='0.9'>0.9배</option>
          <option value='0.8'>0.8배</option>
          <option value='0.7'>0.7배</option>
          <option value='0.5'>0.5배</option>
      </select>
      <input type='button' value='Pause/Resume' onClick='pause()'>
      <div class='jerry_tts_body'>
  '''

$ cat Dockerfile
FROM python:3.7-slim

# Copy local code to the container image.
ENV APP_HOME /app
WORKDIR $APP_HOME
COPY . ./

# Install production dependencies.
RUN pip install Flask gunicorn
RUN pip install google-api-python-client google-auth-httplib2 google-auth-oauthlib
RUN pip install oauth2client
RUN pip install google-cloud
RUN pip install google-cloud-texttospeech
RUN pip install argparse

EXPOSE 5001

# Run the web service on container startup. Here we use the gunicorn
# webserver, with one worker process and 8 threads.
# For environments with multiple CPU cores, increase the number of workers
# to be equal to the cores available.
CMD exec gunicorn --bind :5001 --workers 1 --threads 8 --timeout 0 tts-read-google-docs5:app

$ cat 02.test_local.sh
if [ "$1" == "build" ];
then

mkdir -p build
cd build
cp ../tts-read-google-docs5.py .
cp ../api-project-249965614499-33c4f9f41469.json .
cp ../Dockerfile .
docker build -t "google-docs-tts" .

fi

docker run --name jerry -d -p 5001:5001 google-docs-tts

sleep 1
docker logs -f jerry &

sleep 1
echo Enter to stop...
read a

docker stop jerry
docker rm jerry

$ cat 01.build.sh
mkdir -p build
cd build
cp ../tts-read-google-docs5.py .
cp ../api-project-249965614499-33c4f9f41469.json .
cp ../Dockerfile .
cp ../favicon.ico .
gcloud builds submit --project api-project-249965614499 --tag gcr.io/api-project-249965614499/google-docs-tts

$ cat 03.deploy.sh
gcloud run deploy --project api-project-249965614499 --image gcr.io/api-project-249965614499/google-docs-tts --platform managed --region asia-northeast1 --port 5001 --allow-unauthenticated google-docs-tts