revert to wav file recording

app.py

@@ -41,6 +41,11 @@ def synthesise_audio(text, forward_params=None):
         raise ValueError("Error: El texto es demasiado largo. Por favor, limita tu entrada a 100 caracteres.")
 
     speech = synthesiser(text, forward_params={"speaker_embeddings": speaker_embedding})
+    
+    
+    sf.write("speech.wav", speech["audio"], samplerate=speech["sampling_rate"])
+    return "speech.wav"
+
     # sf.write("speech.wav", speech["audio"], samplerate=speech["sampling_rate"])
     # return "speech.wav"
     
@@ -59,21 +64,33 @@ def synthesise_audio(text, forward_params=None):
     #return speech["audio"]
 
     #return audio
+    
+    # Ensure audio is a numpy array
+    #if isinstance(speech["audio"], int):
+    #    audio = np.array([speech["audio"]])
+    #else:
+    #    audio = speech["audio"]
 
 
     # Create an in-memory buffer to store the audio data
-    audio_buffer = io.BytesIO()
+    #print("Creating in-memory buffer")
+    #audio_buffer = io.BytesIO()
 
     # Write the audio data to the in-memory buffer
-    sf.write(audio_buffer, speech["audio"], samplerate=speech["sampling_rate"], format="WAV")
+    #print("Writing audio data to in-memory buffer")
+    #sf.write(audio_buffer, speech["audio"], samplerate=speech["sampling_rate"], format="WAV")
 
     # Move the buffer cursor to the beginning of the buffer
-    audio_buffer.seek(0)
+    #audio_buffer.seek(0)
 
     # Read the audio data from the in-memory buffer into a numpy array
-    audio, sr = sf.read(audio_buffer)
+    #print("Reading audio data from in-memory buffer")
+    #audio, sr = sf.read(audio_buffer)
+    #print("Audio data read from in-memory buffer, returning audio data and sample rate")
+    # Ensure audio is a numpy array before returning
+    #audio = np.array(audio)
 
-    return audio, sr    
+    #return audio, sr     
   
     
 
@@ -85,8 +102,8 @@ input_text = gr.Textbox(lines=10, label="Enter text here")
 
 demo = gr.Interface(fn=synthesise_audio,
                     inputs=input_text,
-                    #outputs="audio",
-                    outputs = gr.Audio(type="numpy"),
+                    outputs="audio",
+                    #outputs = gr.Audio(type="numpy"),
                     description="----- manuai Text To Speech generator -----",
                     allow_flagging = False)