Spaces:
Running
Running
File size: 9,773 Bytes
8b1f711 e9402b5 8573823 8b1f711 8573823 89ed0ae e9402b5 8573823 e9402b5 8573823 50b6262 8b1f711 8573823 e9402b5 8573823 e9402b5 8573823 e9402b5 8573823 89ed0ae 8573823 e9402b5 8573823 e9402b5 8573823 50b6262 8573823 e9402b5 8573823 e9402b5 8573823 e9402b5 8573823 e9402b5 8573823 e9402b5 8573823 e9402b5 8573823 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 |
import os
import requests
import numpy as np
import streamlit as st
from src.retrieval import STANDARD_QUERIES
from src.content.common import (
MODEL_NAMES,
AUDIO_SAMPLES_W_INSTRUCT,
AGENT_DIALOGUE_STATES,
reset_states,
update_voice_instruction_state,
init_state_section,
header_section,
sidebar_fragment,
successful_example_section,
audio_attach_dialogue,
retrive_response_with_ui
)
API_BASE_URL = os.getenv('API_BASE_URL')
LLM_NO_AUDIO_PROMPT_TEMPLATE = """{user_question}"""
LLM_PROMPT_TEMPLATE = """User asked a question about the audio clip.
## User Question
{user_question}
{audio_information_prompt}Please reply to user's question with a friendly, accurate, and helpful answer."""
AUDIO_INFO_TEMPLATE = """Here are some information about this audio clip.
## Audio Information
{audio_information}
However, the audio analysis may or may not contain relevant information to the user question, please only reply the user with the relevant information.
"""
AUDIO_ANALYSIS_STATUS = "MERaLiON-AudioLLM Analysis"
AG_CONVERSATION_STATES = dict(
ag_messages=[],
ag_model_messages=[],
ag_visited_query_indices=[],
)
def bottom_input_section():
bottom_cols = st.columns([0.03, 0.03, 0.91, 0.03])
with bottom_cols[0]:
st.button(
':material/delete:',
disabled=st.session_state.disprompt,
on_click=lambda: reset_states(AGENT_DIALOGUE_STATES)
)
with bottom_cols[1]:
if st.button(":material/add:", disabled=st.session_state.disprompt):
audio_attach_dialogue(
audio_array_state="ag_audio_array",
audio_base64_state="ag_audio_base64",
restore_state=AG_CONVERSATION_STATES
)
with bottom_cols[2]:
if chat_input := st.chat_input(
placeholder="Instruction...",
disabled=st.session_state.disprompt,
on_submit=lambda: st.session_state.update(disprompt=True)
):
st.session_state.new_prompt = chat_input
with bottom_cols[3]:
uploaded_voice = st.audio_input(
label="voice_instruction",
label_visibility="collapsed",
disabled=st.session_state.disprompt,
on_change=lambda: st.session_state.update(
disprompt=True,
on_record_voice_instruction=True
),
key='voice_instruction'
)
if uploaded_voice and st.session_state.on_record_voice_instruction:
voice_bytes = uploaded_voice.read()
update_voice_instruction_state(voice_bytes)
st.session_state.on_record_voice_instruction = False
def _prepare_final_prompt_with_ui(one_time_prompt):
if st.session_state.ag_audio_array.shape[0] == 0:
return LLM_NO_AUDIO_PROMPT_TEMPLATE.format(user_question=one_time_prompt)
with st.spinner("Searching appropriate querys..."):
response = requests.get(
f"{API_BASE_URL}retrieve_relevant_docs",
params={"user_question": one_time_prompt}
)
relevant_query_indices = response.json()
if len(st.session_state.ag_messages) <= 2:
relevant_query_indices.append(0)
relevant_query_indices = list(
set(relevant_query_indices).difference(st.session_state.ag_visited_query_indices)
)
st.session_state.ag_visited_query_indices.extend(relevant_query_indices)
if not relevant_query_indices:
return LLM_PROMPT_TEMPLATE.format(
user_question=one_time_prompt,
audio_information_prompt=""
)
audio_info = []
with st.status(AUDIO_ANALYSIS_STATUS, expanded=False) as status:
for i, standard_idx in enumerate(relevant_query_indices):
new_label = (
f"{AUDIO_ANALYSIS_STATUS}: "
f"{STANDARD_QUERIES[standard_idx]['ui_text']} "
f"({i+1}/{len(relevant_query_indices)})"
)
status.update(label=new_label, state="running")
error_msg, warnings, response = retrive_response_with_ui(
model_name=MODEL_NAMES["audiollm"]["vllm_name"],
text_input=STANDARD_QUERIES[standard_idx]["query_text"],
array_audio_input=st.session_state.ag_audio_array,
base64_audio_input=st.session_state.ag_audio_base64,
prefix=f"**{STANDARD_QUERIES[standard_idx]['ui_text']}**: ",
stream=True,
show_warning=i==0
)
audio_info.append(STANDARD_QUERIES[standard_idx]["response_prefix_text"] + response)
st.session_state.ag_messages[-1]["process"].append({
"error": error_msg,
"warnings": warnings,
"content": response
})
status.update(label=AUDIO_ANALYSIS_STATUS, state="complete")
audio_information_prompt = AUDIO_INFO_TEMPLATE.format(
audio_information="\n".join(audio_info)
)
return LLM_PROMPT_TEMPLATE.format(
user_question=one_time_prompt,
audio_information_prompt=audio_information_prompt
)
def conversation_section():
chat_message_container = st.container(height=480)
if st.session_state.ag_audio_array.size:
with chat_message_container.chat_message("user"):
st.audio(st.session_state.ag_audio_array, format="audio/wav", sample_rate=16000)
for message in st.session_state.ag_messages:
with chat_message_container.chat_message(name=message["role"]):
if message.get("error"):
st.error(message["error"])
for warning_msg in message.get("warnings", []):
st.warning(warning_msg)
if process := message.get("process", []):
with st.status(AUDIO_ANALYSIS_STATUS, expanded=False, state="complete"):
for proc in process:
if proc.get("error"):
st.error(proc["error"])
for proc_warning_msg in proc.get("warnings", []):
st.warning(proc_warning_msg)
if proc.get("content"):
st.write(proc["content"])
if message.get("content"):
st.write(message["content"])
with st._bottom:
bottom_input_section()
if (not st.session_state.new_prompt) and (not st.session_state.new_vi_base64):
return
one_time_prompt = st.session_state.new_prompt
one_time_vi_array = st.session_state.new_vi_array
one_time_vi_base64 = st.session_state.new_vi_base64
st.session_state.update(
new_prompt="",
new_vi_array=np.array([]),
new_vi_base64="",
)
with chat_message_container.chat_message("user"):
if one_time_vi_base64:
with st.spinner("Transcribing..."):
error_msg, warnings, one_time_prompt = retrive_response_with_ui(
model_name=MODEL_NAMES["audiollm"]["vllm_name"],
text_input="Write out the dialogue as text.",
array_audio_input=one_time_vi_array,
base64_audio_input=one_time_vi_base64,
stream=False,
normalise_response=True
)
else:
error_msg, warnings = "", []
st.write(one_time_prompt)
st.session_state.ag_messages.append({
"role": "user",
"error": error_msg,
"warnings": warnings,
"content": one_time_prompt
})
with chat_message_container.chat_message("assistant"):
assistant_message = {"role": "assistant", "process": []}
st.session_state.ag_messages.append(assistant_message)
final_prompt = _prepare_final_prompt_with_ui(one_time_prompt)
llm_response_prefix = f"**{MODEL_NAMES['llm']['ui_name']}**: "
error_msg, warnings, response = retrive_response_with_ui(
model_name=MODEL_NAMES["llm"]["vllm_name"],
text_input=final_prompt,
array_audio_input=st.session_state.ag_audio_array,
base64_audio_input="",
prefix=llm_response_prefix,
stream=True,
history=st.session_state.ag_model_messages,
show_warning=False
)
assistant_message.update({
"error": error_msg,
"warnings": warnings,
"content": response
})
pure_response = response.replace(llm_response_prefix, "")
st.session_state.ag_model_messages.extend([
{"role": "user", "content": final_prompt},
{"role": "assistant", "content": pure_response}
])
st.session_state.disprompt=False
st.rerun(scope="app")
def agent_page():
init_state_section()
header_section(
component_name="Chatbot",
description=""" It is implemented by <strong>connecting multiple AI models</strong>,
offers more flexibility, and supports <strong>multi-round</strong> conversation.""",
concise_description=""" It is implemented by connecting multiple AI models and
support <strong>multi-round</strong> conversation.""",
icon="👥"
)
with st.sidebar:
sidebar_fragment()
audio_sample_names = [name for name in AUDIO_SAMPLES_W_INSTRUCT.keys() if "Paral" in name]
successful_example_section(
audio_sample_names,
audio_array_state="ag_audio_array",
audio_base64_state="ag_audio_base64",
restore_state=AG_CONVERSATION_STATES
)
conversation_section() |