基于Python开发语音助手并实现用户输入存储的技术问询

Hey Lea, let's tackle your three technical questions step by step, plus fix up your existing code to make the storage work smoothly:

1. Dictionary or DataFrame? Which to Choose? #

• List of dictionaries is perfect for lightweight, in-memory storage if you only need basic append operations and don’t plan on complex data tasks later. It’s simple, low-overhead, and easy to get started with.
• Pandas DataFrame is better if you want to:
  • Filter stories by timestamp, user, or content down the line
  • Export stories to CSV/Excel for family sharing
  • Run quick analytics (like counting stories added per week)

For your family story keeper use case, I’d recommend starting with a list of dicts (easy to implement) and converting it to a DataFrame later if you need more features. Or go straight for a DataFrame if you know you’ll want those advanced capabilities.

2. How to Split Input into Timestamp & User Input Columns? #

You just need to capture the exact time when the user finishes speaking, then pair it with their story. Use datetime.datetime.now() to generate a timestamp, and format it for readability.

Here’s how to structure each entry:

from datetime import datetime

# Create a single story entry with timestamp and content
story_entry = {
    "timestamp": datetime.now().strftime("%Y-%m-%d %H:%M:%S"),  # Formatted as "YYYY-MM-DD HH:MM:SS"
    "user_input": captured_story  # The story you get from takeCommand()
}

3. How to Add a New Row on Each New User Input? #

• If using a list of dicts: Initialize an empty list at the start (e.g., story_log = []), then append each story_entry to this list every time a user shares a story.
• If using a DataFrame: Initialize an empty DataFrame with your columns, then use pd.concat() (since df.append() is deprecated) to add new rows, or assign directly with df.loc[len(df)] = new_entry.

Fixed & Updated Code #

Let’s integrate all this into your existing code, fixing gaps in getstory() and append():

import speech_recognition as sr
from datetime import datetime
import pandas as pd
import pyttsx3

# Initialize speech engine (you likely had this but it was missing in your snippet)
engine = pyttsx3.init()

def speak(audio):
    engine.say(audio)
    engine.runAndWait()

def wishMe():
    hour = int(datetime.now().hour)
    if hour >= 0 and hour < 12:
        speak("Good Morning!")
    elif hour >= 12 and hour < 18:
        speak("Good Afternoon!")
    else:
        speak("Good Evening!")
    assistantname = "Anna"
    speak("I am your personal storykeeper " + assistantname)
    # Note: Speech might have a tiny delay due to engine processing, but it will follow sequentially

def username():
    speak("I would love to know more about you, so let's start off right away. What's your name?")
    username = takeCommand()
    while username == "None":  # Retry if recognition fails
        username = takeCommand()
    speak("Nice to meet you " + username + ".")
    print("Nice to meet you " + username + ".")
    return username  # Return name to link with stories

def storykeepergoal():
    speak("Would you like to know more about my goal to conserve your family's stories?")
    query = takeCommand().lower()
    if "yes" in query:
        msg = "Happy that you are interested to learn more about me. You've probably experienced the moment at a family's birthday party when it's all about telling stories from when you were little, or a funny story about a vacation. As life goes on, those stories might be forgotten or cannot be told anymore. This is where I come in - by telling me stories about your favourite moments in life, I can store these forever. Your family can do the same thing. And if you would like to hear a story that your family members told me, you can tune in."
        speak(msg)
        print(msg)
    else:
        speak("Alright, no problem.")
        print("Alright, no problem.")

def takeCommand():
    r = sr.Recognizer()
    with sr.Microphone() as source:
        print("Listening...")
        r.pause_threshold = 1
        audio = r.listen(source)
    try:
        print("Recognizing...")
        query = r.recognize_google(audio, language='en-GB')
        print(f"User said: {query}\n")
    except Exception as e:
        print(e)
        speak("My apologies, I did not understand that. Could you please say that again?")
        print("My apologies, I did not understand that. Could you please say that again?")
        return "None"
    return query

# Initialize storage (pick one option)
# Option 1: Lightweight list of dicts
story_log = []

# Option 2: Feature-rich DataFrame
# story_df = pd.DataFrame(columns=["timestamp", "user_input", "username"])

def getstory(username):
    speak("I am curious to hear what memories and stories you have about your family. Please tell me your favourite memory with, or about your family.")
    user_story = takeCommand()
    while user_story == "None":  # Ensure we get a valid story
        user_story = takeCommand()
    
    # Create timestamped entry
    timestamp = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
    story_entry = {
        "timestamp": timestamp,
        "user_input": user_story,
        "username": username  # Optional: Track who shared the story
    }
    
    # Add to storage
    story_log.append(story_entry)
    # For DataFrame:
    # global story_df
    # story_df = pd.concat([story_df, pd.DataFrame([story_entry])], ignore_index=True)
    
    speak("Thank you for sharing that story! I've saved it for you.")
    print(f"Story saved at {timestamp}:\n{user_story}\n")

def view_stories():
    # Optional: View all saved stories
    print("=== Saved Family Stories ===")
    # For list of dicts
    for idx, entry in enumerate(story_log, 1):
        print(f"\nStory {idx}:")
        print(f"Timestamp: {entry['timestamp']}")
        print(f"Shared by: {entry['username']}")
        print(f"Content: {entry['user_input']}")
    
    # For DataFrame:
    # print(story_df.to_string(index=False))

# Run the assistant
if __name__ == "__main__":
    wishMe()
    user_name = username()
    storykeepergoal()
    getstory(user_name)
    
    # Example: Add another story later
    # speak("Would you like to share another story?")
    # if "yes" in takeCommand().lower():
    #     getstory(user_name)
    
    # View all saved stories
    view_stories()

Key Improvements: #

• Added speech engine initialization (missing in your original snippet)
• Made username() return the user’s name to link stories with contributors
• Fixed getstory() to capture valid input, create timestamped entries, and save them
• Added a view_stories() function to check stored entries
• Included retry logic in takeCommand() to handle recognition failures
• Offered both storage options so you can pick what fits your needs

内容的提问来源于stack exchange，提问作者Lea