Update app.py

app.py

@@ -1,95 +1,76 @@
 import os
+import time
 import requests
 import gradio as gr
-import asyncio
-import aiohttp
 from huggingface_hub import login
 from dotenv import load_dotenv
 
 load_dotenv()
 image_API="https://api-inference.huggingface.co/models/stabilityai/stable-diffusion-2"
-translation_API="https://api-inference.huggingface.co/models/facebook/nllb-200-distilled-600M"
-whisper_API="https://api-inference.huggingface.co/models/openai/whisper-medium"
-txt_API="https://api-inference.huggingface.co/models/openai-community/gpt2"
-HF_TOKEN=os.getenv("HF_TOKEN")
+translation_API="https://api-inference.huggingface.co/models/facebook/nllb-200-distilled-1.3B"
+whisper_API="https://api-inference.huggingface.co/models/openai/whisper-large-v3"
+
+HF_TOKEN=os.getenv("HF_Token")
 login(HF_TOKEN)
 headers={"Authorization":f"Bearer {HF_TOKEN}"}
 
-async def query_api(url, payload=None, data=None):
-    async with aiohttp.ClientSession() as session:
-      for attempts in range(5):
-        try:
-          async with session.post(url, headers=headers, json=payload, data=data) as response:
-            if response.status == 200:
-              return await response.json()
-            print("API Error {response.status}, retrying....")
-        except Exception as e:
-          print(f"Error: {e}")
-        await asyncio.sleep(10)
-    return None
 
-async def query_whisper(audio_path):
+def query_whisper(audio_path):
   with open(audio_path,"rb") as audio:
     data=audio.read()
-  return await query_api(whisper_API,data=data)
+    response=requests.post(whisper_API,headers=headers,data=data)
+    return response.json()
 
-async def query_translation(text):
+def query_translation(text):
   max_retries=5
   delay=10
   payload={"inputs":text,"parameters":{"src_lang":"ta_Taml","tgt_lang":'eng_Latn'}}
-  response=await query_api(translation_API,json=payload)
-  if response and isinstance(response,list) and len(response)>0:
-      return response[0]["translation_text"]
-  elif response and "translation_text" in resopnse:
-    return response["translation_text"]
-  return "Translation Error"
+  for attempt in range(max_retries):
+    response=requests.post(translation_API,headers=headers,json=payload)
+    result= response.json()
+    if isinstance(result,list) and len(result)>0:
+      return {"translated_text":result[0]["translation_text"]}
+    elif isinstance(result,dict) and "translation_text" in result:
+      return {"translated_text":result["translation_text"]}
+    time.sleep(delay)
+  return {"error":"Translation API failed","response":result}
 
-async def query_image(prompt):
+def query_image(prompt):
+  max_retries=5
+  delay=10
   payload={"inputs":prompt}
-  response=await query_api(image_API,json=payload)
-  if response:
-    image_path='generated_image.png'
-    with open(image_path,'wb') as f:
-      f.write(response.content)
-    return image_path
-  return None
-
-async def query_text_generation(prompt):
-  payload={"inputs":f"give me a short story about {prompt}"}
-  response=await query_api(txt_API,json=payload)
-  if response and isinstance(response, list) and len(response) > 0:
-    return response[0].get("generated_text", "Text Generation Error")
-  elif response and "generated_text" in response:
-    return response["generated_text"]
+  for attempt in range(max_retries):
+    response=requests.post(image_API,headers=headers,json=payload)
+    if response.status_code == 200 and response.headers.get('Content-Type', '').startswith('image/'):
+      image_path='generated_image.png'
+      with open(image_path,'wb') as f:
+        f.write(response.content)
+      return image_path
+    time.sleep(delay)
   return None
 
 
-
-
-async def process_audio(audio_path):
+def process_audio(audio_path):
   if not audio_path:
-    return None,"Audio not provided",None,None
+    return None
   try:
-    transcription=await query_whisper(audio_path)
+    transcription=query_whisper(audio_path)
     tamil_text=transcription.get("text","Transcription error")
-
-    translation_task=query_translation(tamil_text)
-    image_task=query_image(translated_text)
-    story_task=query_text_generation(translated_text)
-
-    translated_text,image_path,story=await asyncio.gather(translation_task,image_task,story_task)
-    return tamil_text,translated_text,image_path,story
+    translation=query_translation(tamil_text)
+    translated_text=translation.get("translated_text","Translation error")
+    image_path=query_image(translated_text)
+    return tamil_text,translated_text,image_path
   except Exception as e:
     return None,str(e),None
 
 iface=gr.Interface(
-    fn=lambda x:asyncio.run(process_audio(x)),
+    fn=process_audio,
     inputs=gr.Audio(type="filepath",label="Upload Audio"),
     outputs=[
         gr.Textbox(label="Tamil Text"),
         gr.Textbox(label="English Translation"),
-        gr.Image(label="Generated Image"),
-        gr.Textbox(label="Story")
+        gr.Image(label="Generated Image")
+        
     ],
     title="Speech-to-Image Generation"
 )