Disable audio generation

main.py

@@ -1,4 +1,4 @@
-from fastapi import FastAPI, HTTPException, UploadFile, File
+from fastapi import FastAPI, HTTPException, FileResponse
 from fastapi.middleware.cors import CORSMiddleware
 from fastapi.responses import JSONResponse
 from utils.palmoil_classification import AfroPalmModel
@@ -69,8 +69,9 @@ async def predict(image_request: ImageRequest):
             logging.debug(f"Prediction: {prediction}, Confidence: {confidence*100:.2f}%")
 
             # Generate audio
-            audio_generation = AudioGeneration(prediction=prediction, confidence=confidence*100, language=image_request.lang)
-            translated_text = audio_generation.ghanaian_language_translator()
+            # audio_generation = AudioGeneration(prediction=prediction, confidence=confidence*100, language=image_request.lang)
+            # translated_text = audio_generation.ghanaian_language_translator()
+            # translated_text_audiofile = audio_generation.text_to_audio(translated_text)
 
         else:
             logging.info("Image is not a red palm oil")
@@ -86,8 +87,10 @@ async def predict(image_request: ImageRequest):
         return {
             "status": "success",
             "result": prediction,
-            "confidence": f"{confidence*100:.2f}"
+            "confidence": f"{confidence*100:.2f}",
+
         }
+            # "audio":  FileResponse(path='final_result.wav', media_type="audio/mpeg", filename="final_result.wav")
 
     except Exception as e:
         logging.error(e)

requirements.txt

@@ -7,4 +7,4 @@ torch
 torchvision
 opencv-python
 tensorflow
-roboflow
+roboflow

utils/audio_generation.py

@@ -1,5 +1,7 @@
 import requests
 import logging
+from gradio_client import Client
+
 
 class AudioGeneration:
     """
@@ -64,4 +66,23 @@ class AudioGeneration:
         translated_text = response['data'][0]
         logging.debug(f"Translated text:{translated_text}")
 
-        return translated_text
+        return translated_text
+    
+    def text_to_audio(self,text):
+        """
+        Convert the translated text to audio
+        :param text: translated text
+        :return: audio file
+        """
+        logging.info("Converting text to audio")
+
+        client = Client("https://softwarearoma-ghanalanguageaudiosynthesizer.hf.space/--replicas/0t7ah/")
+        result = client.predict(
+                self.selected_languages[self.language],
+                text,	
+                api_name="/predict"
+        )
+        
+        logging.debug(f"Audio file: {result[0]}")
+
+        return result[0]