Update app.py
Browse files
app.py
CHANGED
@@ -1,14 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
1 |
import whisper
|
2 |
import os
|
3 |
from gtts import gTTS
|
4 |
import gradio as gr
|
5 |
from groq import Groq
|
6 |
-
from datetime import datetime
|
7 |
-
from IPython.display import display, HTML
|
8 |
|
9 |
# Load a smaller Whisper model for faster processing
|
10 |
try:
|
11 |
-
model = whisper.load_model("tiny"
|
12 |
except Exception as e:
|
13 |
print(f"Error loading Whisper model: {e}")
|
14 |
model = None
|
@@ -89,43 +93,6 @@ def chatbot(audio):
|
|
89 |
print(f"Error in chatbot function: {e}")
|
90 |
return "Sorry, there was an error processing your request.", None, chat_history
|
91 |
|
92 |
-
# Inject custom CSS for background and styling
|
93 |
-
def set_background():
|
94 |
-
display(HTML(f'''
|
95 |
-
<style>
|
96 |
-
body {{
|
97 |
-
background-image: url("https://github.com/abdullahzunorain/voice-to-voice-Chatbot/blob/main/image_1");
|
98 |
-
background-size: cover;
|
99 |
-
background-position: center;
|
100 |
-
background-repeat: no-repeat;
|
101 |
-
color: white;
|
102 |
-
font-family: Arial, sans-serif;
|
103 |
-
}}
|
104 |
-
.gradio-container {{
|
105 |
-
background-color: rgba(0, 0, 0, 0.6);
|
106 |
-
padding: 20px;
|
107 |
-
border-radius: 8px;
|
108 |
-
box-shadow: 0 4px 8px rgba(0, 0, 0, 0.2);
|
109 |
-
}}
|
110 |
-
h1, h2, p, .gradio-label {{
|
111 |
-
color: #FFD700; /* Gold color for labels and titles */
|
112 |
-
}}
|
113 |
-
.gradio-button {{
|
114 |
-
background-color: #FFD700;
|
115 |
-
color: black;
|
116 |
-
border-radius: 4px;
|
117 |
-
font-weight: bold;
|
118 |
-
}}
|
119 |
-
.gradio-input {{
|
120 |
-
background-color: rgba(255, 255, 255, 0.9);
|
121 |
-
border-radius: 4px;
|
122 |
-
}}
|
123 |
-
</style>
|
124 |
-
'''))
|
125 |
-
|
126 |
-
# Display custom background styling
|
127 |
-
set_background()
|
128 |
-
|
129 |
# Gradio interface for real-time interaction with chat history display
|
130 |
iface = gr.Interface(
|
131 |
fn=chatbot,
|
@@ -137,161 +104,41 @@ iface = gr.Interface(
|
|
137 |
live=True,
|
138 |
title="Stylish Audio Chatbot with Groq API",
|
139 |
description="Upload your audio, and the chatbot will transcribe and respond to it with a synthesized response.",
|
140 |
-
theme="default"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
141 |
)
|
142 |
|
143 |
-
# Launch the Gradio app
|
144 |
if __name__ == "__main__":
|
145 |
-
iface.launch(
|
146 |
-
|
147 |
-
|
148 |
-
|
149 |
-
|
150 |
-
|
151 |
-
#----------------------------------------------------------------------------------------------------------------------------------------------------------
|
152 |
-
|
153 |
-
|
154 |
-
|
155 |
-
|
156 |
-
|
157 |
-
|
158 |
-
|
159 |
-
# import whisper
|
160 |
-
# import os
|
161 |
-
# from gtts import gTTS
|
162 |
-
# import gradio as gr
|
163 |
-
# from groq import Groq
|
164 |
-
# from datetime import datetime
|
165 |
-
|
166 |
-
# # Load a smaller Whisper model for faster processing
|
167 |
-
# try:
|
168 |
-
# model = whisper.load_model("tiny")
|
169 |
-
# except Exception as e:
|
170 |
-
# print(f"Error loading Whisper model: {e}")
|
171 |
-
# model = None
|
172 |
-
|
173 |
-
# # Set up Groq API client using environment variable
|
174 |
-
# GROQ_API_TOKEN = os.getenv("GROQ_API")
|
175 |
-
# if not GROQ_API_TOKEN:
|
176 |
-
# raise ValueError("Groq API token is missing. Set 'GROQ_API' in your environment variables.")
|
177 |
-
# client = Groq(api_key=GROQ_API_TOKEN)
|
178 |
-
|
179 |
-
# # Initialize the chat history
|
180 |
-
# chat_history = []
|
181 |
-
|
182 |
-
# # Function to get the LLM response from Groq with timeout handling
|
183 |
-
# def get_llm_response(user_input, role="detailed responder"):
|
184 |
-
# prompt = f"As an expert, provide a detailed and knowledgeable response: {user_input}" if role == "expert" else \
|
185 |
-
# f"As a good assistant, provide a clear, concise, and helpful response: {user_input}" if role == "good assistant" else \
|
186 |
-
# f"Provide a thorough and detailed response: {user_input}"
|
187 |
-
|
188 |
-
# try:
|
189 |
-
# chat_completion = client.chat.completions.create(
|
190 |
-
# messages=[{"role": "user", "content": user_input}],
|
191 |
-
# model="llama3-8b-8192", # Replace with your desired model
|
192 |
-
# timeout=20 # Increased timeout to 20 seconds
|
193 |
-
# )
|
194 |
-
# return chat_completion.choices[0].message.content
|
195 |
-
# except Exception as e:
|
196 |
-
# print(f"Error during LLM response retrieval: {e}")
|
197 |
-
# return "Sorry, there was an error retrieving the response. Please try again."
|
198 |
-
|
199 |
-
# # Function to convert text to speech using gTTS
|
200 |
-
# def text_to_speech(text):
|
201 |
-
# try:
|
202 |
-
# tts = gTTS(text)
|
203 |
-
# output_audio = "output_audio.mp3"
|
204 |
-
# tts.save(output_audio)
|
205 |
-
# return output_audio
|
206 |
-
# except Exception as e:
|
207 |
-
# print(f"Error generating TTS: {e}")
|
208 |
-
# return None
|
209 |
-
|
210 |
-
# # Main chatbot function to handle audio input and output with chat history
|
211 |
-
# def chatbot(audio):
|
212 |
-
# if not model:
|
213 |
-
# return "Error: Whisper model is not available.", None, chat_history
|
214 |
-
|
215 |
-
# if not audio:
|
216 |
-
# return "No audio provided. Please upload a valid audio file.", None, chat_history
|
217 |
-
|
218 |
-
# try:
|
219 |
-
# # Step 1: Transcribe the audio using Whisper
|
220 |
-
# result = model.transcribe(audio)
|
221 |
-
# user_text = result.get("text", "")
|
222 |
-
# if not user_text.strip():
|
223 |
-
# return "Could not understand the audio. Please try speaking more clearly.", None, chat_history
|
224 |
-
|
225 |
-
# # Get current timestamp
|
226 |
-
# timestamp = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
|
227 |
-
|
228 |
-
# # Display transcription in chat history
|
229 |
-
# chat_history.append((timestamp, "User", user_text))
|
230 |
-
|
231 |
-
# # Step 2: Get LLM response from Groq
|
232 |
-
# response_text = get_llm_response(user_text)
|
233 |
-
|
234 |
-
# # Step 3: Convert the response text to speech
|
235 |
-
# output_audio = text_to_speech(response_text)
|
236 |
-
|
237 |
-
# # Append the latest interaction to the chat history
|
238 |
-
# chat_history.append((timestamp, "Chatbot", response_text))
|
239 |
-
|
240 |
-
# # Format the chat history for display with timestamps and clear labels
|
241 |
-
# formatted_history = "\n".join([f"[{time}] {speaker}: {text}" for time, speaker, text in chat_history])
|
242 |
-
|
243 |
-
# return formatted_history, output_audio, chat_history
|
244 |
-
|
245 |
-
# except Exception as e:
|
246 |
-
# print(f"Error in chatbot function: {e}")
|
247 |
-
# return "Sorry, there was an error processing your request.", None, chat_history
|
248 |
-
|
249 |
-
# # Gradio interface for real-time interaction with chat history display
|
250 |
-
# iface = gr.Interface(
|
251 |
-
# fn=chatbot,
|
252 |
-
# inputs=gr.Audio(type="filepath"),
|
253 |
-
# outputs=[
|
254 |
-
# gr.Textbox(label="Chat History"), # Display chat history
|
255 |
-
# gr.Audio(type="filepath", label="Response Audio"),
|
256 |
-
# ],
|
257 |
-
# live=True,
|
258 |
-
# title="Stylish Audio Chatbot with Groq API",
|
259 |
-
# description="Upload your audio, and the chatbot will transcribe and respond to it with a synthesized response.",
|
260 |
-
# theme="default",
|
261 |
-
# css='''
|
262 |
-
# body {
|
263 |
-
# background-image: url("https://raw.githubusercontent.com/username/repository/main/path/to/your-image.png");
|
264 |
-
# background-size: cover;
|
265 |
-
# background-position: center;
|
266 |
-
# background-repeat: no-repeat;
|
267 |
-
# color: white;
|
268 |
-
# font-family: Arial, sans-serif;
|
269 |
-
# }
|
270 |
-
# .gradio-container {
|
271 |
-
# background-color: rgba(0, 0, 0, 0.6);
|
272 |
-
# padding: 20px;
|
273 |
-
# border-radius: 8px;
|
274 |
-
# box-shadow: 0 4px 8px rgba(0, 0, 0, 0.2);
|
275 |
-
# }
|
276 |
-
# h1, h2, p, .gradio-label {
|
277 |
-
# color: #FFD700; /* Gold color for labels and titles */
|
278 |
-
# }
|
279 |
-
# .gradio-button {
|
280 |
-
# background-color: #FFD700;
|
281 |
-
# color: black;
|
282 |
-
# border-radius: 4px;
|
283 |
-
# font-weight: bold;
|
284 |
-
# }
|
285 |
-
# .gradio-input {
|
286 |
-
# background-color: rgba(255, 255, 255, 0.9);
|
287 |
-
# border-radius: 4px;
|
288 |
-
# }
|
289 |
-
# '''
|
290 |
-
# )
|
291 |
-
|
292 |
-
# # Launch the Gradio app
|
293 |
-
# if __name__ == "__main__":
|
294 |
-
# iface.launch()
|
295 |
|
296 |
|
297 |
|
|
|
1 |
+
|
2 |
+
|
3 |
+
|
4 |
+
|
5 |
+
|
6 |
import whisper
|
7 |
import os
|
8 |
from gtts import gTTS
|
9 |
import gradio as gr
|
10 |
from groq import Groq
|
11 |
+
from datetime import datetime
|
|
|
12 |
|
13 |
# Load a smaller Whisper model for faster processing
|
14 |
try:
|
15 |
+
model = whisper.load_model("tiny")
|
16 |
except Exception as e:
|
17 |
print(f"Error loading Whisper model: {e}")
|
18 |
model = None
|
|
|
93 |
print(f"Error in chatbot function: {e}")
|
94 |
return "Sorry, there was an error processing your request.", None, chat_history
|
95 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
96 |
# Gradio interface for real-time interaction with chat history display
|
97 |
iface = gr.Interface(
|
98 |
fn=chatbot,
|
|
|
104 |
live=True,
|
105 |
title="Stylish Audio Chatbot with Groq API",
|
106 |
description="Upload your audio, and the chatbot will transcribe and respond to it with a synthesized response.",
|
107 |
+
theme="default",
|
108 |
+
css='''
|
109 |
+
body {
|
110 |
+
background-image: url("https://raw.githubusercontent.com/username/repository/main/path/to/your-image.png");
|
111 |
+
background-size: cover;
|
112 |
+
background-position: center;
|
113 |
+
background-repeat: no-repeat;
|
114 |
+
color: white;
|
115 |
+
font-family: Arial, sans-serif;
|
116 |
+
}
|
117 |
+
.gradio-container {
|
118 |
+
background-color: rgba(0, 0, 0, 0.6);
|
119 |
+
padding: 20px;
|
120 |
+
border-radius: 8px;
|
121 |
+
box-shadow: 0 4px 8px rgba(0, 0, 0, 0.2);
|
122 |
+
}
|
123 |
+
h1, h2, p, .gradio-label {
|
124 |
+
color: #FFD700; /* Gold color for labels and titles */
|
125 |
+
}
|
126 |
+
.gradio-button {
|
127 |
+
background-color: #FFD700;
|
128 |
+
color: black;
|
129 |
+
border-radius: 4px;
|
130 |
+
font-weight: bold;
|
131 |
+
}
|
132 |
+
.gradio-input {
|
133 |
+
background-color: rgba(255, 255, 255, 0.9);
|
134 |
+
border-radius: 4px;
|
135 |
+
}
|
136 |
+
'''
|
137 |
)
|
138 |
|
139 |
+
# Launch the Gradio app
|
140 |
if __name__ == "__main__":
|
141 |
+
iface.launch()
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
142 |
|
143 |
|
144 |
|