Spaces:
Sleeping
Sleeping
File size: 3,337 Bytes
8666fb2 9e32d2f 8666fb2 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 | import gradio as gr
import pandas as pd
import os
import tempfile
import re
from transformers import pipeline
# -------- Load Whisper Model (HF) --------
asr = pipeline(
"automatic-speech-recognition",
model="openai/whisper-small"
)
# -------- Speech to Text --------
def transcribe_audio(file_path):
result = asr(file_path)
return result["text"]
# -------- Simple NLP Extraction (Rule-based) --------
def extract_fields(text):
data = {
"Name": "",
"Phone": "",
"Product": "",
"Budget": "",
"Location": "",
"Intent": ""
}
text_lower = text.lower()
# Name extraction
match = re.search(r"my name is (\w+)", text_lower)
if match:
data["Name"] = match.group(1).title()
# Phone extraction
phone_match = re.search(r"\b\d{10}\b", text)
if phone_match:
data["Phone"] = phone_match.group()
# Budget extraction
if "lakh" in text_lower or "budget" in text_lower:
data["Budget"] = text
# Product (basic)
if "car" in text_lower:
data["Product"] = "Car"
elif "bike" in text_lower:
data["Product"] = "Bike"
else:
data["Product"] = text[:50]
# Location (basic guess)
if "in" in text_lower:
data["Location"] = text
# Intent
if "buy" in text_lower or "interested" in text_lower:
data["Intent"] = "Hot Lead"
else:
data["Intent"] = "Cold Lead"
return data
# -------- Main Processing --------
def process_audio(audio_file):
if audio_file is None:
return "No audio provided", pd.DataFrame(), None
try:
file_path = audio_file # Gradio gives filepath
# Step 1: Transcription
text = transcribe_audio(file_path)
# Step 2: Extraction
data = extract_fields(text)
# Step 3: DataFrame
df = pd.DataFrame([data])
# Step 4: Save Excel
excel_path = os.path.join(tempfile.gettempdir(), "crm_output.xlsx")
df.to_excel(excel_path, index=False)
return text, df, excel_path
except Exception as e:
return f"Error: {str(e)}", pd.DataFrame(), None
# -------- UI --------
with gr.Blocks() as app:
gr.Markdown("# 🎙️ AI Voice to CRM Auto Filler ")
with gr.Tabs():
# 🎤 Record
with gr.Tab("🎤 Record Inquiry"):
mic_input = gr.Audio(
sources=["microphone"],
type="filepath",
label="Record Audio"
)
btn1 = gr.Button("Process Recording")
# 📁 Upload
with gr.Tab("📁 Upload Voice"):
file_input = gr.Audio(
sources=["upload"],
type="filepath",
label="Upload Audio File"
)
btn2 = gr.Button("Process File")
# Outputs
transcript_output = gr.Textbox(label="Transcription")
table_output = gr.Dataframe(label="Extracted CRM Data")
download_btn = gr.File(label="Download Excel")
# Actions
btn1.click(
fn=process_audio,
inputs=mic_input,
outputs=[transcript_output, table_output, download_btn]
)
btn2.click(
fn=process_audio,
inputs=file_input,
outputs=[transcript_output, table_output, download_btn]
)
# Launch
app.launch()
|