11.11实时字幕推荐

import speech_recognition as sr
from nltk.corpus import stopwords
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import linear_kernel

# 初始化语音识别器
r = sr.Recognizer()

# 获取音频流并进行语音识别
def get_transcript(audio_source):
    with audio_source as source:
        audio = r.listen(source)
    try:
        transcript = r.recognize_google(audio)
        return transcript
    except sr.UnknownValueError:
        return "无法识别语音"
    except sr.RequestError:
        return "无法请求结果"

# 推荐相关字幕
def recommend_subtitle(transcript, subtitles):
    tfidf = TfidfVectorizer(stop_words='english')
    tfidf_matrix = tfidf.fit_transform(subtitles + [transcript])
    cosine_sim = linear_kernel(tfidf_matrix[-1:], tfidf_matrix[:-1]).flatten()
    related_indices = cosine_sim.argsort()[:-5:-1]
    return [subtitles[i] for i in related_indices]

# 示例使用
audio_source = sr.Microphone()
transcript = get_transcript(audio_source)
subtitles = ["科技新闻", "体育赛事", "娱乐八卦", "经济分析"]
recommended_subtitles = recommend_subtitle(transcript, subtitles)
print("推荐字幕:", recommended_subtitles)