Update audio_pipe.py

audio_pipe.py

@@ -2,6 +2,8 @@ import json
 import os
 from pathlib import Path
 from typing import List, Tuple
+import tempfile
+import soundfile as sf
 
 import numpy as np
 import torch
@@ -148,5 +150,12 @@ class SpeechToSpeechPipeline():
             wav, sr = TTSHubInterface.get_prediction(
                 self.tts_task, self.tts_model, self.tts_generator, tts_sample
             )
+        temp_name = ""
+        with tempfile.NamedTemporaryFile(suffix=".wav") as tmp_output_file:
+            sf.write(tmp_output_file, wav.detach().cpu().numpy(), sr)
+            tmp_output_file.seek(0)
+            temp_name = tmp_output_file.name
+            print(temp_name)
 
-        return wav, sr, [text]
+        # return wav, sr, [text]
+        return temp_name