def speech_recognition(job, url):
ACCESS_KEY_ID = access_key_id()
SECRET_ACCESS_KEY = secret_access_key()
bucket_name = s3_bucket_name()
region = s3_region()
transcribe = boto3.client('transcribe',
region_name=region,
aws_access_key_id=ACCESS_KEY_ID,
aws_secret_access_key=SECRET_ACCESS_KEY)
job_name = str(job)
job_uri = str(url)
path = urllib.parse.urlparse(url).path
ext = os.path.splitext(path)[1].lstrip('.')
transcribe.start_transcription_job(OutputBucketName=bucket_name,
TranscriptionJobName=job_name,
Media={'MediaFileUri': job_uri},
MediaFormat=ext,
LanguageCode='en-US')
while True:
status = transcribe.get_transcription_job(
TranscriptionJobName=job_name)
if status['TranscriptionJob']['TranscriptionJobStatus'] in [
'COMPLETED', 'FAILED'
]:
break
print("Not ready yet...")
time.sleep(5)
return status
def extract_image(x):
ACCESS_KEY_ID = access_key_id()
SECRET_ACCESS_KEY = secret_access_key()
bucket = s3_bucket_name()
region = s3_region()
# connect to s3 client and list objects in bucket:
# s3 = boto3.client('s3', region_name = region, aws_access_key_id = ACCESS_KEY_ID, aws_secret_access_key = SECRET_ACCESS_KEY)
# bucket_response = s3.list_objects_v2(Bucket=bucket)
# for i in bucket_response['Contents']:
# y = i['Key']
# print(y)
rekognition = boto3.client('rekognition',
region_name=region,
aws_access_key_id=ACCESS_KEY_ID,
aws_secret_access_key=SECRET_ACCESS_KEY)
response = rekognition.detect_text(
Image={'S3Object': {
'Bucket': bucket,
'Name': x
}})
resp_str = ""
for resp in response['TextDetections']:
if 'ParentId' in resp:
resp_str += resp['DetectedText'] + ' '
key = ['Text']
val = [resp_str]
keyval = dict(zip(key, val))
return keyval
def upload_image(x):
ACCESS_KEY_ID = access_key_id()
SECRET_ACCESS_KEY = secret_access_key()
bucket_name = s3_bucket_name()
region = s3_region()
s3 = boto3.client('s3', region_name = region, aws_access_key_id=ACCESS_KEY_ID, aws_secret_access_key=SECRET_ACCESS_KEY)
filename = x
s3.upload_file(filename, bucket_name, filename)
key = ['Message']
val = [x + ' uploaded']
keyval = dict(zip(key,val))
return keyval
def upload_audio(URL):
ACCESS_KEY_ID = access_key_id()
SECRET_ACCESS_KEY = secret_access_key()
bucket_name = s3_bucket_name()
region = s3_region()
keyname = s3_key()
file_name = URL.split('/')[-1]
urllib.request.urlopen(URL).read()
s3 = boto3.client('s3',
region_name=region,
aws_access_key_id=ACCESS_KEY_ID,
aws_secret_access_key=SECRET_ACCESS_KEY)
s3.upload_file(file_name, bucket_name, keyname.format(file_name))
key = ['Message']
val = [file_name + ' uploaded']
keyval = dict(zip(key, val))
return keyval
def extract_image(URL):
ACCESS_KEY_ID = access_key_id()
SECRET_ACCESS_KEY = secret_access_key()
region = s3_region()
res = requests.get(URL) #Fetching URL response...
byteObj = res.content #Converting URL object to Byte object
img = Image.open(BytesIO(byteObj))
rekognition = boto3.client('rekognition',
region_name=region,
aws_access_key_id=ACCESS_KEY_ID,
aws_secret_access_key=SECRET_ACCESS_KEY)
response = rekognition.detect_text(Image={'Bytes': byteObj})
resp_txt = []
for i in response['TextDetections']:
# if 'ParentId' in i :
if 'Id' in i and 'ParentId' not in i:
resp_txt.append(i['DetectedText'])
resp_key = []
for k in range(len(resp_txt)):
resp_key.append(k)
my_dict = dict(zip(resp_key, resp_txt))
for key, val in my_dict.items():
# if re.search(r'\w+@\w+', val):
if re.search(r'(^[a-zA-Z0-9_.+-]+@[a-zA-Z0-9-]+\.[a-zA-Z0-9-.]+$)',
val):
my_dict['email'] = my_dict.pop(key)
break
for key, val in my_dict.items():
# if re.search(r'^[1-9]\d{2}-\d{3}-\d{4}', val):
if re.search(r'^\+?\d[\d -]{8,12}\d', val) or re.search(
r'^(\d{3}[-\.\s]??\d{3}[-\.\s]??\d{4}|\(\d{3}\)\s*\d{3}[-\.\s]??\d{4}|\d{3}[-\.\s]??\d{4})',
val
) or re.search(r'^[1-9]\d{2}-\d{3}-\d{4}', val) or re.search(
r'^(\+\d{1,2}\s)?\(?\d{3}\)?[\s.-]\d{3}[\s.-]\d{4}', val
) or re.search(
r'^\s*(?:\+?(\d{1,3}))?[-. (]*(\d{3})[-. )]*(\d{3})[-. ]*(\d{4})(?: *x(\d+))?\s*',
val) or re.search(
r'^(?:(?:\+|0{0,2})91(\s*[\-]\s*)?|[0]?)?[789]\d{9}',
val):
my_dict['phone'] = my_dict.pop(key)
break
# nouns = {x.name().split('.', 1)[0] for x in wn.all_synsets('n')}
# for key,val in my_dict.items():
# if val in nouns:
# my_dict['name'] = my_dict.pop(key)
# break
for key, val in my_dict.items():
if nlp(val):
my_dict['name'] = my_dict.pop(key)
break
for key, val in my_dict.items():
if re.search(
r'^(https?:\/\/(?:www\.|(?!www))[a-zA-Z0-9][a-zA-Z0-9-]+[a-zA-Z0-9]\.[^\s]{2,}|www\.[a-zA-Z0-9][a-zA-Z0-9-]+[a-zA-Z0-9]\.[^\s]{2,}|https?:\/\/(?:www\.|(?!www))[a-zA-Z0-9]+\.[^\s]{2,}|www\.[a-zA-Z0-9]+\.[^\s]{2,})',
val):
my_dict['url'] = my_dict.pop(key)
break
lst = ['email', 'phone', 'name', 'url']
dict1 = {}
unknown = []
for m, n in my_dict.items():
if m not in lst:
unknown.append(n)
newdict = {k: my_dict[k] for k in lst if k in my_dict}
dict1['results'] = unknown
dict2 = {**newdict, **dict1}
return dict2