Spaces:
Paused
Paused
avinash
commited on
Commit
Β·
a50f676
1
Parent(s):
0a81205
revert
Browse files
app.py
CHANGED
|
@@ -1,13 +1,34 @@
|
|
| 1 |
import gradio as gr
|
|
|
|
|
|
|
|
|
|
| 2 |
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
gr.Markdown("This Space showcases the nari-labs/Dia-1.6B model, served by the fal-ai API. Sign in with your Hugging Face account to use this API.")
|
| 7 |
-
button = gr.LoginButton("Sign in")
|
| 8 |
-
gr.load("models/nari-labs/Dia-1.6B", accept_token=button, provider="fal-ai")
|
| 9 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
|
| 12 |
if __name__ == "__main__":
|
| 13 |
-
|
|
|
|
| 1 |
import gradio as gr
|
| 2 |
+
from asr import transcribe_audio
|
| 3 |
+
from llm_agent import get_llm_reply
|
| 4 |
+
from tts import generate_speech
|
| 5 |
|
| 6 |
+
def voice_agent(audio):
|
| 7 |
+
if audio is None:
|
| 8 |
+
return "No audio received!", "...", None
|
|
|
|
|
|
|
|
|
|
| 9 |
|
| 10 |
+
# ASR
|
| 11 |
+
user_text = transcribe_audio(audio)
|
| 12 |
+
|
| 13 |
+
# LLM
|
| 14 |
+
bot_reply = get_llm_reply(user_text)
|
| 15 |
+
|
| 16 |
+
# TTS
|
| 17 |
+
audio_reply = generate_speech(bot_reply)
|
| 18 |
+
|
| 19 |
+
return user_text, bot_reply, audio_reply
|
| 20 |
|
| 21 |
+
ui = gr.Interface(
|
| 22 |
+
fn=voice_agent,
|
| 23 |
+
inputs=gr.Audio(type="filepath", label="ποΈ Upload Voice (.wav or .mp3)"),
|
| 24 |
+
outputs=[
|
| 25 |
+
gr.Textbox(label="π Transcribed Text"),
|
| 26 |
+
gr.Textbox(label="π€ LLM Reply"),
|
| 27 |
+
gr.Audio(type="filepath", label="π AI Voice Response")
|
| 28 |
+
],
|
| 29 |
+
title="VoiceFreight AI v1.0",
|
| 30 |
+
description="Upload a voice file. Get a spoken reply from an AI freight agent!"
|
| 31 |
+
)
|
| 32 |
|
| 33 |
if __name__ == "__main__":
|
| 34 |
+
ui.launch()
|