import os import uuid from django.http import FileResponse from rest_framework import status from rest_framework.response import Response from rest_framework.generics import CreateAPIView from TTS.api import TTS from .serializers import TextToSpeechSerializer class TextToSpeechCreateView(CreateAPIView): serializer_class = TextToSpeechSerializer def create(self, request, *args, **kwargs): serializer = self.get_serializer(data=request.data) if serializer.is_valid(): text = serializer.validated_data.get("text") speaker_wav = serializer.validated_data.get("speaker_wav") language = serializer.validated_data.get("language") output_filename = f"output_{uuid.uuid4()}.wav" try: # Save the uploaded speaker file to a temporary location speaker_file_path = os.path.join("/tmp", speaker_wav.name) with open(speaker_file_path, "wb") as destination: for chunk in speaker_wav.chunks(): destination.write(chunk) # Generate speech using tts.tts_to_file tts = TTS("tts_models/multilingual/multi-dataset/xtts_v2", gpu=False) tts.tts_to_file(text=text, file_path=output_filename, speaker_wav=speaker_file_path, language=language) # Define a function to delete the output file def file_iterator(file_name): with open(file_name, 'rb') as f: yield from f # Delete the file after sending it try: os.remove(file_name) except Exception as e: # You might want to log this error pass # Use the file_iterator to create a FileResponse response = FileResponse(file_iterator(output_filename), as_attachment=True, content_type='audio/wav') return response except Exception as e: return Response({"error": str(e)}, status=status.HTTP_500_INTERNAL_SERVER_ERROR) return Response(serializer.errors, status=status.HTTP_400_BAD_REQUEST)