1) Input as video input and receive as flac result 2) Upload flac file to cloud storege 3) Extract text by voice recognition 4) Save video name & text to fire store

mp4_convert.py

from __future__ import print_function  # for compatibility with both python 2 and 3
from subprocess import call  # for calling mplayer and lame
import os  # help with file handling


def check_file_exists(directory, filename, extension):
    path = directory + "/" + filename + extension
    return os.path.isfile(path)


def main(indir, outdir):
    try:
        # check specified folders exist
        if not os.path.exists(indir):
            exit("Error: Input directory \'" + indir + "\' does not exist. (try prepending './')")
        if not os.path.exists(outdir):
            exit("Error: Output directory \'" + outdir + "\' does not exist.")
        if not os.access(outdir, os.W_OK):
            exit("Error: Output directory \'" + outdir + "\' is not writeable.")

        print("[{0}/*.mp4] --> [{1}/*.mp3]".format(indir, outdir))
        files = []  # files for exporting

        # get a list of all convertible files in the input directory
        filelist = [f for f in os.listdir(indir) if f.endswith(".mp4")]
        for path in filelist:
            basename = os.path.basename(path)
            filename = os.path.splitext(basename)[0]
            files.append(filename)
        # remove files that have already been outputted from the list
        files[:] = [f for f in files if not check_file_exists(outdir, f, ".mp3")]
    except OSError as e:
        exit(e)

    if len(files) == 0:
        exit("Could not find any files to convert that have not already been converted.")

    # convert all unconverted files
    for filename in files:
        print("-- converting {0}/{2}.mp4 to {1}/{2}.mp3 --".format(indir, outdir, filename))
        call(["mplayer", "-novideo", "-nocorrect-pts", "-ao", "pcm:waveheader", indir + "/" + filename + ".mp4"])
        call(["lame", "-v", "audiodump.wav", outdir + "/" + filename + ".mp3"])
        call(['ffmpeg', '-i', '{}/{}.mp3'.format(outdir, filename),
              '-ac', '1', '{}/{}.flac'.format(outdir, filename)])
        call(['rm', '-rf', '{}/{}.mp3'.format(outdir, filename)])
        # ToDO google storage upload!!
        os.remove("audiodump.wav")


# set the default directories and try to get input directories
args = ["./video", "./output"]

main(args[0], args[1])

speech_upload.py

from google.cloud import speech
from google.cloud.speech import enums
from google.cloud.speech import types
from google.cloud import storage
import firebase_admin
from firebase_admin import credentials
from firebase_admin import firestore
import os

# Instantiates a client
cred = credentials.ApplicationDefault()
firebase_admin.initialize_app(cred, {
    'projectId': '{project-id}',
})

db = firestore.client()
client = speech.SpeechClient()
storage_client = storage.Client()
bucket = storage_client.get_bucket('{bucket-name}')

output_dir = 'output'
for item in os.listdir(output_dir):
    print('{} file stroage upload!'.format(item))
    blob = bucket.blob('output/' + item)
    blob.upload_from_filename(os.path.join(output_dir, item))

config = types.RecognitionConfig(
    encoding=enums.RecognitionConfig.AudioEncoding.FLAC,
    language_code='ko-KR',
    sample_rate_hertz=44100,
)

bucket_url = 'gs://{bucket-url}'
# file_names = ['AT-cm_224754895-480.flac', 'AT-cm_226859162-480.flac']
file_names = os.listdir('output')

doc_ref = db.collection(u'label')

for name in file_names:
    audio = types.RecognitionAudio(
        uri=bucket_url + name,
    )

    response = client.long_running_recognize(
        config=config, audio=audio
    )

    op_result = response.result()

    for result in op_result.results:
        for alternative in result.alternatives:
            print(alternative.transcript)
            print(alternative.confidence)

            doc_ref.document().set({
                u'subscription': str.encode(alternative.transcript),
                u'video_name': str.encode(name)
            })