Spaces:
Build error
Build error
Enhance process_input and create_demo functions in app.py to improve multimodal input handling, including better formatting for user messages and integration of TextStreamer for text response generation.
Browse files
app.py
CHANGED
@@ -1,6 +1,6 @@
|
|
1 |
import gradio as gr
|
2 |
import torch
|
3 |
-
from transformers import Qwen2_5OmniModel, Qwen2_5OmniProcessor
|
4 |
from qwen_omni_utils import process_mm_info
|
5 |
import soundfile as sf
|
6 |
import tempfile
|
@@ -51,7 +51,16 @@ def process_input(image, audio, video, text, chat_history, voice_type, enable_au
|
|
51 |
if isinstance(item, list) and len(item) == 2:
|
52 |
user_msg, bot_msg = item
|
53 |
if bot_msg is not None: # Only add complete message pairs
|
54 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
55 |
conversation.append({"role": "assistant", "content": bot_msg})
|
56 |
else:
|
57 |
# Initialize chat history if it's not a list
|
@@ -78,14 +87,19 @@ def process_input(image, audio, video, text, chat_history, voice_type, enable_au
|
|
78 |
)
|
79 |
inputs = inputs.to(model.device).to(model.dtype)
|
80 |
|
81 |
-
# Generate response
|
82 |
if enable_audio_output:
|
83 |
voice_type_value = VOICE_OPTIONS.get(voice_type, "Chelsie")
|
84 |
text_ids, audio = model.generate(
|
85 |
**inputs,
|
86 |
use_audio_in_video=False, # Set to False to avoid audio processing issues
|
87 |
return_audio=True,
|
88 |
-
spk=voice_type_value
|
|
|
|
|
|
|
|
|
|
|
89 |
)
|
90 |
|
91 |
# Save audio to temporary file
|
@@ -100,7 +114,12 @@ def process_input(image, audio, video, text, chat_history, voice_type, enable_au
|
|
100 |
text_ids = model.generate(
|
101 |
**inputs,
|
102 |
use_audio_in_video=False, # Set to False to avoid audio processing issues
|
103 |
-
return_audio=False
|
|
|
|
|
|
|
|
|
|
|
104 |
)
|
105 |
audio_path = None
|
106 |
|
@@ -111,17 +130,20 @@ def process_input(image, audio, video, text, chat_history, voice_type, enable_au
|
|
111 |
clean_up_tokenization_spaces=False
|
112 |
)[0]
|
113 |
|
114 |
-
# Clean up text response
|
115 |
text_response = text_response.strip()
|
|
|
|
|
|
|
116 |
|
117 |
# Format user message for chat history display
|
118 |
user_message_for_display = str(text) if text is not None else ""
|
119 |
if image is not None:
|
120 |
-
user_message_for_display = (user_message_for_display
|
121 |
if audio is not None:
|
122 |
-
user_message_for_display = (user_message_for_display
|
123 |
if video is not None:
|
124 |
-
user_message_for_display = (user_message_for_display
|
125 |
|
126 |
# If empty, provide a default message
|
127 |
if not user_message_for_display.strip():
|
@@ -168,7 +190,12 @@ def create_demo():
|
|
168 |
# Chat interface
|
169 |
with gr.Row():
|
170 |
with gr.Column(scale=3):
|
171 |
-
chatbot = gr.Chatbot(
|
|
|
|
|
|
|
|
|
|
|
172 |
with gr.Accordion("Advanced Options", open=False):
|
173 |
voice_type = gr.Dropdown(
|
174 |
choices=list(VOICE_OPTIONS.keys()),
|
@@ -185,9 +212,11 @@ def create_demo():
|
|
185 |
with gr.TabItem("Text Input"):
|
186 |
text_input = gr.Textbox(
|
187 |
placeholder="Type your message here...",
|
188 |
-
label="Text Input"
|
|
|
|
|
189 |
)
|
190 |
-
text_submit = gr.Button("Send Text")
|
191 |
|
192 |
with gr.TabItem("Multimodal Input"):
|
193 |
with gr.Row():
|
@@ -205,9 +234,10 @@ def create_demo():
|
|
205 |
)
|
206 |
additional_text = gr.Textbox(
|
207 |
placeholder="Additional text message...",
|
208 |
-
label="Additional Text"
|
|
|
209 |
)
|
210 |
-
multimodal_submit = gr.Button("Send Multimodal Input")
|
211 |
|
212 |
clear_button = gr.Button("Clear Chat")
|
213 |
|
|
|
1 |
import gradio as gr
|
2 |
import torch
|
3 |
+
from transformers import Qwen2_5OmniModel, Qwen2_5OmniProcessor, TextStreamer
|
4 |
from qwen_omni_utils import process_mm_info
|
5 |
import soundfile as sf
|
6 |
import tempfile
|
|
|
51 |
if isinstance(item, list) and len(item) == 2:
|
52 |
user_msg, bot_msg = item
|
53 |
if bot_msg is not None: # Only add complete message pairs
|
54 |
+
# Convert display format back to processable format
|
55 |
+
processed_msg = user_msg
|
56 |
+
if "[Image]" in user_msg:
|
57 |
+
processed_msg = {"type": "text", "text": user_msg.replace("[Image]", "").strip()}
|
58 |
+
if "[Audio]" in user_msg:
|
59 |
+
processed_msg = {"type": "text", "text": user_msg.replace("[Audio]", "").strip()}
|
60 |
+
if "[Video]" in user_msg:
|
61 |
+
processed_msg = {"type": "text", "text": user_msg.replace("[Video]", "").strip()}
|
62 |
+
|
63 |
+
conversation.append({"role": "user", "content": processed_msg})
|
64 |
conversation.append({"role": "assistant", "content": bot_msg})
|
65 |
else:
|
66 |
# Initialize chat history if it's not a list
|
|
|
87 |
)
|
88 |
inputs = inputs.to(model.device).to(model.dtype)
|
89 |
|
90 |
+
# Generate response with streaming
|
91 |
if enable_audio_output:
|
92 |
voice_type_value = VOICE_OPTIONS.get(voice_type, "Chelsie")
|
93 |
text_ids, audio = model.generate(
|
94 |
**inputs,
|
95 |
use_audio_in_video=False, # Set to False to avoid audio processing issues
|
96 |
return_audio=True,
|
97 |
+
spk=voice_type_value,
|
98 |
+
max_new_tokens=512,
|
99 |
+
do_sample=True,
|
100 |
+
temperature=0.7,
|
101 |
+
top_p=0.9,
|
102 |
+
streamer=TextStreamer(processor, skip_prompt=True)
|
103 |
)
|
104 |
|
105 |
# Save audio to temporary file
|
|
|
114 |
text_ids = model.generate(
|
115 |
**inputs,
|
116 |
use_audio_in_video=False, # Set to False to avoid audio processing issues
|
117 |
+
return_audio=False,
|
118 |
+
max_new_tokens=512,
|
119 |
+
do_sample=True,
|
120 |
+
temperature=0.7,
|
121 |
+
top_p=0.9,
|
122 |
+
streamer=TextStreamer(processor, skip_prompt=True)
|
123 |
)
|
124 |
audio_path = None
|
125 |
|
|
|
130 |
clean_up_tokenization_spaces=False
|
131 |
)[0]
|
132 |
|
133 |
+
# Clean up text response by removing system/user messages
|
134 |
text_response = text_response.strip()
|
135 |
+
text_response = text_response.split("assistant")[-1].strip()
|
136 |
+
if text_response.startswith(":"):
|
137 |
+
text_response = text_response[1:].strip()
|
138 |
|
139 |
# Format user message for chat history display
|
140 |
user_message_for_display = str(text) if text is not None else ""
|
141 |
if image is not None:
|
142 |
+
user_message_for_display = (user_message_for_display + " " if user_message_for_display.strip() else "") + "[Image]"
|
143 |
if audio is not None:
|
144 |
+
user_message_for_display = (user_message_for_display + " " if user_message_for_display.strip() else "") + "[Audio]"
|
145 |
if video is not None:
|
146 |
+
user_message_for_display = (user_message_for_display + " " if user_message_for_display.strip() else "") + "[Video]"
|
147 |
|
148 |
# If empty, provide a default message
|
149 |
if not user_message_for_display.strip():
|
|
|
190 |
# Chat interface
|
191 |
with gr.Row():
|
192 |
with gr.Column(scale=3):
|
193 |
+
chatbot = gr.Chatbot(
|
194 |
+
height=600,
|
195 |
+
show_label=False,
|
196 |
+
avatar_images=["👤", "🤖"],
|
197 |
+
bubble_full_width=False,
|
198 |
+
)
|
199 |
with gr.Accordion("Advanced Options", open=False):
|
200 |
voice_type = gr.Dropdown(
|
201 |
choices=list(VOICE_OPTIONS.keys()),
|
|
|
212 |
with gr.TabItem("Text Input"):
|
213 |
text_input = gr.Textbox(
|
214 |
placeholder="Type your message here...",
|
215 |
+
label="Text Input",
|
216 |
+
autofocus=True,
|
217 |
+
container=False,
|
218 |
)
|
219 |
+
text_submit = gr.Button("Send Text", variant="primary")
|
220 |
|
221 |
with gr.TabItem("Multimodal Input"):
|
222 |
with gr.Row():
|
|
|
234 |
)
|
235 |
additional_text = gr.Textbox(
|
236 |
placeholder="Additional text message...",
|
237 |
+
label="Additional Text",
|
238 |
+
container=False,
|
239 |
)
|
240 |
+
multimodal_submit = gr.Button("Send Multimodal Input", variant="primary")
|
241 |
|
242 |
clear_button = gr.Button("Clear Chat")
|
243 |
|