app.py

import time
import pandas as pd
import numpy as np
import tiktoken
from openai.embeddings_utils import get_embedding, cosine_similarity
import translators as ts
import speech_recognition as sr
import gradio as gr
import openai
import subprocess


openai.api_key = "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"

df = pd.read_csv("absher_user_guide_with_embeddings_1k.csv")
df["embedding"] = df.embedding.apply(eval).apply(np.array)

def search_services(df, user_input, n=3, pprint=False):
    product_embedding = get_embedding(
        user_input,
        engine="text-embedding-ada-002"
    )
    df["similarity"] = df.embedding.apply(lambda x: cosine_similarity(x, product_embedding))

    results = (
        df.sort_values("similarity", ascending=False)
        .head(n)
        .combined.str.replace("Title: ", "")
        .str.replace("; Content:", ": ")
    )
    results = (
        df.sort_values("similarity", ascending=False)
        .head(n)
        .Steps_Arabic
    ).values[0]

    if pprint:
        for r in results:
            print(r[:200])
            print()
    return results

def listen():
    r = sr.Recognizer()

    with sr.Microphone() as source:
        print("May I help you?")
        audio = r.listen(source)

    try:
        text = r.recognize_google(audio, language="ar-AE", show_all=False)
        print("You said: " + text)
    except sr.UnknownValueError:
        print("Speech Recognition could not understand audio")
    except sr.RequestError as e:
        print("Could not request results from Speech Recognition service; {0}".format(e))
    
    return text




def transcribe(audio_filepath):
    harvard = sr.AudioFile(audio_filepath)
    r = sr.Recognizer()
    with harvard as source:
        audio = r.record(source)
    try:
        text = r.recognize_google(audio, language="ar-AE", show_all=False)
        print("You said: " + text)
    except sr.UnknownValueError:
        print("Speech Recognition could not understand audio")
    except sr.RequestError as e:
        print("Could not request results from Speech Recognition service; {0}".format(e))
    return text

def run(audio_filepath):
    text = transcribe(audio_filepath)
    results = search_services(df, ts.translate_text(text), n=1)
    return results

demo = gr.Interface(
    fn=run, 
    inputs=gr.Audio(source="microphone", type="filepath"), 
    outputs="text").launch()

if __name__ == "__main__":
    demo.launch()

# prompt = listen()
# results = search_services(df, ts.translate_text(prompt), n=3)

# print(results)