This is a draft cheat sheet. It is a work in progress and is not finished yet.
Get started
|
create a resource and choose Visual RecognitionMarkup in the AI category |
use Lite Plan and copy your given API key and your URL |
install Python Version 3 |
install these packages with pip: opencv-python/ ibm-watson |
Face Recognition code
import cv2
from ibm_watson import VisualRecognitionV3
from ibm_cloud_sdk_core.authenticators import IAMAuthenticator
import json
import sys
authenticator = IAMAuthenticator('YOUR API KEY')
service = VisualRecognitionV3(
version='2020-02-26',
authenticator=authenticator
)
service.set_service_url('YOUR URL')
cap = cv2.VideoCapture(0)
key = ''
while( key != 'q' ):
ret, frame = cap.read()
cv2.imshow('frame', frame)
intkey = cv2.waitKey(2)
if( intkey > 0 ):
key = chr(intkey)
else:
key = ''
if key == 'd' :
cv2.imwrite('upload.jpg', frame)
with open('upload.jpg','rb') as file:
classes = service.classify(images_file=file,threshold='0.6').get_result()
print(json.dumps(classes, indent=2))
|
|
|
IBM Watson in Java
IamOptions iamoptions = new IamOptions.Builder().apiKey("YOUR API KEY").build();
SpeechToText stt = new SpeechToText(iamoptions);
stt.setEndPoint("https://stream-fra.watsonplatform.net/speech-to-text/api");
File audio = new File("src/resources/sample1.wav");
RecognizeOptions options = new RecognizeOptions.Builder()
.audio(audio)
.contentType(HttpMediaType.AUDIO_WAV)
.build();
SpeechRecognitionResults transcript = stt.recognize(options).execute();
System.out.println(transcript);
taText.setText(transcript.getResults().get(0).getAlternatives().get(0).getTranscript());
|
Speech to Text in Python
$ pip install --upgrade watson-developer-cloud
from speech_sentiment_python.recorder import Recorder
recorder = Recorder(“speech.wav”)
recorder.record_to_file()
def transcribe_audio(path_to_audio_file):
username = os.environ.get(“USERNAME”)
password = os.environ.get(“PASSWORD”)
speech_to_text = SpeechToText(username=username,
password=password)
with open(join(dirname(__file__), path_to_audio_file), ‘rb’) as
audio_file:
return speech_to_text.recognize(audio_file,
content_type=’audio/wav’)
|
|
