{"id":957,"date":"2018-08-04T13:47:29","date_gmt":"2018-08-04T11:47:29","guid":{"rendered":"http:\/\/pythonprogramming.altervista.org\/?p=957"},"modified":"2018-08-27T07:54:40","modified_gmt":"2018-08-27T05:54:40","slug":"python-and-speech-recognition","status":"publish","type":"post","link":"https:\/\/pythonprogramming.altervista.org\/python-and-speech-recognition\/","title":{"rendered":"Python and speech recognition"},"content":{"rendered":"

What did you say?<\/h2>\n

\"\"<\/p>\n

As I often talked about text-to speech, why we do non talk about speech recognition? As also\u00a0https:\/\/realpython.com\/<\/a> talked about this topic, I want to share a little code here to give a sample of its possible utilization.<\/p>\n

import time\r\nimport speech_recognition as sr\r\n\r\n\r\ndef elaborate_recording(recognizer, microphone):\r\n    if not isinstance(recognizer, sr.Recognizer):\r\n        raise TypeError(\"A problem with the recognizer\")\r\n    if not isinstance(microphone, sr.Microphone):\r\n        raise TypeError(\"A problem with the microphone\")\r\n    with microphone as source:\r\n        recognizer.adjust_for_ambient_noise(source)\r\n        audio = recognizer.listen(source)\r\n    elaboration = {\r\n        \"success\": True,\r\n        \"error\": None,\r\n        \"transcription\": None\r\n    }\r\n    try:\r\n        elaboration[\"transcription\"] = recognizer.recognize_google(audio)\r\n    except sr.RequestError:\r\n        elaboration[\"success\"] = False\r\n        elaboration[\"error\"] = \"API not available\"\r\n    except sr.UnknownValueError:\r\n        elaboration[\"error\"] = \"I did not understand\"\r\n\r\n    return elaboration\r\n\r\n\r\nif __name__ == \"__main__\":\r\n    print(\"I start the recognizer and the mic.. start talking\")\r\n    recognizer = sr.Recognizer()\r\n    microphone = sr.Microphone()\r\n    print(\"Dici qualcosa adesso \/ Say something now ... pause 3 second ... continue talking\")\r\n    time.sleep(3)\r\n    print(\"NOW!\")\r\n    guess = elaborate_recording(recognizer, microphone)\r\n    print(guess[\"transcription\"])\r\n<\/pre>\n