diff --git a/AiChatterbot.py b/AiChatterbot.py index 37572e0..5098323 100644 --- a/AiChatterbot.py +++ b/AiChatterbot.py @@ -8,21 +8,17 @@ import pyttsx3 from tkinter import * from tkinter import messagebox import speech_recognition as sr -from tkinter import Text, Scrollbar, Entry, Button -import nltk -from nltk.tokenize import word_tokenize -from nltk.corpus import stopwords -from nltk.stem import WordNetLemmatizer - -# Constants for audio recording -framer = 16000 # Sample rate -num_samples = 2000 # Sample points +from tkinter import Text, Entry, Button + +framer = 16000 +num_samples = 2000 channels = 1 sampwidth = 2 -audio_file = 'myvoices.wav' # Specify the path to save the recorded audio file -APP_ID = '73927317' -API_KEY = '3jGcj5fLma64CtB3tTEuLcei' -SECRET_KEY = 'qm8gPCF7DSKpqatx5ZQ8e4OvNLmgdYcG' # Add your own API credentials here +APP_ID = '78542969' +API_KEY = 'B37LWVxLfoHixHx52OrNCPij' +SECRET_KEY = 'bkNnbC37K3Ac6gI8kQ8phG1h5RIivMwj' +client = AipSpeech(APP_ID, API_KEY, SECRET_KEY) +audio_file = '../pythonProject6/myvoices.wav' # Class for handling audio recording and saving class Speak(): @@ -39,16 +35,14 @@ class Speak(): stream = pa.open(format=paInt16, channels=channels, rate=framer, input=True, frames_per_buffer=num_samples) my_buf = [] t_end = time.time() + 5 - print('Speak now...') + print('识别中...') while time.time() < t_end: string_audio_data = stream.read(num_samples) my_buf.append(string_audio_data) - print('Recording ended') + print('识别完成') self.save_wave_file(audio_file, my_buf) stream.close() -client = AipSpeech('73927317', '3jGcj5fLma64CtB3tTEuLcei', 'qm8gPCF7DSKpqatx5ZQ8e4OvNLmgdYcG') - class RobotSay(): def __init__(self): self.engine = pyttsx3.init() @@ -58,12 +52,17 @@ class RobotSay(): self.engine.runAndWait() class ReadWav(): + def __init__(self): + self.APP_ID = '78542969' + self.API_KEY = 'B37LWVxLfoHixHx52OrNCPij' + self.SECRET_KEY = 'bkNnbC37K3Ac6gI8kQ8phG1h5RIivMwj' + self.client = AipSpeech(self.APP_ID, self.API_KEY, self.SECRET_KEY) def get_file_content(self, filePath): with open(filePath, 'rb') as fp: return fp.read() def predict(self): - return client.asr(self.get_file_content('myvoices.wav'), 'wav', 16000, { + return client.asr(self.get_file_content('../pythonProject6/myvoices.wav'), 'wav', 16000, { 'dev_pid': 1537, }) @@ -144,16 +143,38 @@ class ChatWindow: port=3306) self.cursor = self.connection.cursor() self.chat_history = [] + + self.voice_button = Button(self.master, text="语音对话", command=self.start_voice_conversation) + self.voice_button.pack() + + self.APP_ID = APP_ID + self.API_KEY = API_KEY + self.SECRET_KEY = SECRET_KEY + self.client = AipSpeech(APP_ID, API_KEY, SECRET_KEY) + ''' nltk.download('punkt') nltk.download('stopwords') nltk.download('wordnet') + ''' def send_message(self): user_input = self.input_field.get() - self.update_chat("你: " + user_input + "\n") response = self.get_qingyunke_response(user_input) + self.update_chat("你: " + user_input + "\n") self.update_chat("菲菲: " + response + "\n") self.speak(response) + self.save_to_database(user_input, response) + def start_voice_conversation(self): + # Implement voice conversation functionality + self.speak("开始语音对话，请说话...") + speak_instance = Speak() + speak_instance.my_record() + text = self.transcribe_audio('../pythonProject6/myvoices.wav') + self.update_chat("你说: " + text + "\n") + response = self.get_qingyunke_response(text) + self.update_chat("菲菲: " + response + "\n") + self.speak(response) + self.save_to_database(text, response) def get_qingyunke_response(self, message): api_url = "http://api.qingyunke.com/api.php?key=free&appid=0&msg=" + message response = requests.get(api_url) @@ -172,10 +193,15 @@ class ChatWindow: self.engine.runAndWait() def transcribe_audio(self, audio_file): - with sr.AudioFile(audio_file) as source: - audio_data = self.recognizer.record(source) - text = self.recognizer.recognize_sphinx(audio_data) - return text + with open(audio_file, 'rb') as f: + audio_data = f.read() + result = self.client.asr(audio_data, 'wav', 16000, { + 'dev_pid': 1537, # Use the model for Mandarin + }) + if 'result' in result: + return result['result'][0] + else: + return "Baidu speech recognition failed" #使用mysql数据库来保存 def save_to_database(self, user_message, bot_response): sql = "INSERT INTO chat_history (user_message, bot_response) VALUES (%s, %s)" @@ -214,20 +240,25 @@ class ChatWindow: self.cursor.close() self.connection.close() #自然语言处理 - def process_user_input(self, user_input): - # 对用户输入进行分词 - tokens = word_tokenize(user_input) - # 去除停用词 - stop_words = set(stopwords.words('english')) - filtered_tokens = [word for word in tokens if word.lower() not in stop_words] - # 词形还原 - lemmatizer = WordNetLemmatizer() - lemmatized_tokens = [lemmatizer.lemmatize(word) for word in filtered_tokens] - # 将处理后的文本重新组合为句子 - processed_input = ' '.join(lemmatized_tokens) - # 返回处理后的文本 - return processed_input -user_input = "Can you tell me about machine learning?" + + #知识图谱模块 + def fetch_knowledge_graph_info(self, query): + # Assuming you are using Google Knowledge Graph API + api_key = 'YOUR_API_KEY' + base_url = 'https://kgsearch.googleapis.com/v1/entities:search' + + params = { + 'query': query, + 'key': api_key, + 'limit': 1 # You can adjust the limit based on response requirements + } + + response = requests.get(base_url, params=params).json() + + if 'itemListElement' in response: + item = response['itemListElement'][0] + if 'result' in item: + return item['result']['name'], item['result']['detailedDescription']['articleBody'] def main(): root = Tk() login_window = LoginWindow(root) diff --git a/myvoices.wav b/myvoices.wav new file mode 100644 index 0000000..1db7195 Binary files /dev/null and b/myvoices.wav differ