add gradio tts & app.py which make gradio and api same route

Dockerfile

@@ -12,6 +12,8 @@ RUN pip install --no-cache-dir --upgrade -r requirements.txt
 COPY --chown=user . /app
 
 EXPOSE 7826 8000 8001
-CMD python3 interface.py & \
-    uvicorn main:app --host 0.0.0.0 --port 8000 & \
-    uvicorn tts.tts:app --host 0.0.0.0 --port 8001
+# CMD python3 interface.py & \
+#     uvicorn main:app --host 0.0.0.0 --port 8000 & \
+#     uvicorn tts.tts:app --host 0.0.0.0 --port 8001
+
+CMD python3 app.py

app.py

@@ -0,0 +1,10 @@
+import uvicorn
+from main import app
+from interface import create_gradio_interface
+import gradio as gr
+
+if __name__ == "__main__":
+    gr_interface = create_gradio_interface()
+    gr_interface.queue(default_concurrency_limit=1)
+    app = gr.mount_gradio_app(app, gr_interface, path="/")
+    uvicorn.run(app, host='0.0.0.0', port=7860, workers=1)

interface.py

@@ -1,7 +1,9 @@
+import os
 import gradio as gr
 import requests
 
-API_BASE_URL = "http://localhost:8000"  # Update this with your actual API base URL if deployed elsewhere
+PORT = 7860
+API_BASE_URL = f"http://localhost:{PORT}" 
 
 # Function: Get nurse response
 def get_nurse_response(user_input, model_name, chat_history):
@@ -71,7 +73,97 @@ def view_ehr_details(view):
     except requests.exceptions.RequestException as e:
         return f"Error: {str(e)}"
 
-# Gradio Interface
+# --------------------------------------------------------------------------------------------
+def call_botnoi_tts(text, speaker, volume, speed):
+    url = f"{API_BASE_URL}/tts/generate_voice_botnoi/"
+    payload = {
+        "text": text,
+        "speaker": speaker,
+        "volume": volume,
+        "speed": speed,
+        "token": os.getenv("BOTNOI_API_TOKEN")
+    }
+
+    response = requests.post(url, json=payload)
+    if response.status_code == 200:
+        return response.content, "output.mp3"
+    else:
+        return f"Error: {response.status_code} - {response.json().get('detail', 'Unknown error')}", None
+
+# Helper function to call VAJA9 API
+def call_vaja9_tts(text, speaker, phrase_break, audiovisual):
+    url = f"{API_BASE_URL}/tts/generate_voice_vaja9/"
+    payload = {
+        "text": text,
+        "speaker": speaker,
+        "phrase_break": phrase_break,
+        "audiovisual": audiovisual
+    }
+
+    response = requests.post(url, json=payload)
+    if response.status_code == 200:
+        return response.content, "output.wav"
+    else:
+        return f"Error: {response.status_code} - {response.json().get('detail', 'Unknown error')}", None
+# --------------------------------------------------------------------------------------------
+
+def gradio_tts_interface():
+    with gr.Tabs() as tabs:
+            # Tab for Botnoi TTS API
+            with gr.TabItem("Botnoi TTS"):
+                gr.Markdown("### Generate Voice with Botnoi API")
+
+                botnoi_text = gr.Textbox(label="Text", placeholder="Enter text to synthesize")
+                botnoi_speaker = gr.Textbox(label="Speaker ID", value="52", placeholder="Default: 52")
+                botnoi_volume = gr.Slider(label="Volume", minimum=0, maximum=100, value=100)
+                botnoi_speed = gr.Slider(label="Speed", minimum=0.5, maximum=2.0, step=0.1, value=1.0)
+
+                botnoi_generate = gr.Button("Generate Audio")
+                botnoi_output = gr.Audio(label="Generated Audio")
+                botnoi_error = gr.Textbox(label="Error", interactive=False, visible=False)
+
+                def generate_botnoi_voice(text, speaker, volume, speed):
+                    result, file_name = call_botnoi_tts(text, speaker, volume, speed)
+                    if file_name:
+                        return gr.update(value=result), ""
+                    else:
+                        return None, result
+
+                botnoi_generate.click(generate_botnoi_voice, 
+                                    inputs=[botnoi_text, botnoi_speaker, botnoi_volume, botnoi_speed], 
+                                    outputs=[botnoi_output, botnoi_error])
+
+            # Tab for VAJA9 TTS API
+            with gr.TabItem("VAJA9 TTS"):
+                gr.Markdown("### Generate Voice with VAJA9 API")
+
+                vaja9_text = gr.Textbox(label="Text", placeholder="Enter text to synthesize")
+                vaja9_speaker = gr.Radio(label="Speaker", choices=["0 - Male", "1 - Female", "2 - Boy", "3 - Girl"], value="1 - Female")
+                vaja9_phrase_break = gr.Radio(label="Phrase Break", choices=["0 - Auto", "1 - None"], value="0 - Auto")
+                vaja9_audiovisual = gr.Radio(label="Audiovisual", choices=["0 - Audio", "1 - Audio + Visual"], value="0 - Audio")
+
+                vaja9_generate = gr.Button("Generate Audio")
+                vaja9_output = gr.Audio(label="Generated Audio")
+                vaja9_error = gr.Textbox(label="Error", interactive=False, visible=False)
+
+                def generate_vaja9_voice(text, speaker, phrase_break, audiovisual):
+                    speaker_id = int(speaker.split(" - ")[0])
+                    phrase_break_id = int(phrase_break.split(" - ")[0])
+                    audiovisual_id = int(audiovisual.split(" - ")[0])
+
+                    result, file_name = call_vaja9_tts(text, speaker_id, phrase_break_id, audiovisual_id)
+                    if file_name:
+                        return gr.update(value=result), ""
+                    else:
+                        return None, result
+
+                vaja9_generate.click(generate_vaja9_voice, 
+                                    inputs=[vaja9_text, vaja9_speaker, vaja9_phrase_break, vaja9_audiovisual], 
+                                    outputs=[vaja9_output, vaja9_error])
+    return tabs
+
+# --------------------------------------------------------------------------------------------
+# Chatbot Interface
 def create_gradio_interface():
     with gr.Blocks() as interface:
         # Title and description
@@ -176,7 +268,6 @@ def create_gradio_interface():
                 inputs=[],
                 outputs=chat_history_output,
             )
-
             send_button.click(
                 fn=view_ehr_details,
                 inputs=[gr.Textbox(value="details", visible=False)],
@@ -189,15 +280,24 @@ def create_gradio_interface():
                 outputs=ehr_prompt_output
             )         
 
+        gr.Markdown(
+                """
+                ---  
+                """
+            )
+    # TTS --------------------------------------------------------------------------------------------
+        gr.Markdown("# Text-to-Speech (TTS) API Test Interface")
+        tts_interface = gradio_tts_interface()
+
         # Footer
         gr.Markdown(
             """
             ---  
-            Developed by **Piang** 🚀  
+            Built With ❤️ by **[Piang](https://github.com/microhum)** 🚀
             Powered by Typhoon v1.5x and OpenThaiGPT Models.  
             """
         )
-
+        
     return interface
 
 # Run the Gradio Interface

main.py

@@ -1,10 +1,13 @@
+import os
 from typing import Optional
+import requests
 import uvicorn
 from llm.basemodel import EHRModel
 from llm.llm import VirtualNurseLLM
-from fastapi import FastAPI
+from fastapi import FastAPI, HTTPException
 from fastapi.middleware.cors import CORSMiddleware
-from fastapi.responses import HTMLResponse
+from fastapi.responses import FileResponse, HTMLResponse
+from pythainlp.tokenize import sent_tokenize
 from pydantic import BaseModel
 from llm.models import model_list, get_model
 import time
@@ -23,6 +26,7 @@ nurse_llm = VirtualNurseLLM(
 #     api_key="dummy"
 # )
 
+
 app = FastAPI()
 
 app.add_middleware(
@@ -51,21 +55,21 @@ class EHRData(BaseModel):
 class ChatHistory(BaseModel):
     chat_history: list
         
-@app.get("/", response_class=HTMLResponse)
-def read_index():
-    return """
-    <!DOCTYPE html>
-    <html>
-    <head>
-        <title>MALI_NURSE API</title>
-    </head>
-    <body>
-        <h1>Welcome to MALI_NURSE API</h1>
-        <p>This is the index page. Use the link below to access the API docs:</p>
-        <a href="/docs">Go to Swagger Docs UI</a>
-    </body>
-    </html>
-    """
+# @app.get("/", response_class=HTMLResponse)
+# def read_index():
+#     return """
+#     <!DOCTYPE html>
+#     <html>
+#     <head>
+#         <title>MALI_NURSE API</title>
+#     </head>
+#     <body>
+#         <h1>Welcome to MALI_NURSE API</h1>
+#         <p>This is the index page. Use the link below to access the API docs:</p>
+#         <a href="/docs">Go to Swagger Docs UI</a>
+#     </body>
+#     </html>
+#     """
 
 @app.get("/history")
 def get_chat_history():
@@ -124,5 +128,10 @@ def nurse_response(user_input: UserInput):
     
     return NurseResponse(nurse_response=response)
 
+# TTS
+from tts.tts import app as tts_app
+
+app.mount("/tts", tts_app)
+
 if __name__ == "__main__":
     uvicorn.run("main:app", host="0.0.0.0", port=8000, reload=True)