add wav/ogg option

app.py

@@ -152,7 +152,7 @@ def inference(text, language, noise_scale, noise_scale_w, length_scale, voice, f
             x_tst_lengths = torch.LongTensor([stn_tst.size(0)])
             audio = model.infer(x_tst, x_tst_lengths, noise_scale=noise_scale,
                                 noise_scale_w=noise_scale_w, length_scale=length_scale)[0][0, 0].data.float().numpy()
-            result = [ipa_text, (hps_vctk.data.sampling_rate, float2pcm(audio))]
+            result = [ipa_text, (hps_vctk.data.sampling_rate, audio)]
     else:
         ipa_text, stn_tst = get_text(text, language, hps_vctk, mode="VITS")
         with torch.no_grad():
@@ -161,9 +161,11 @@ def inference(text, language, noise_scale, noise_scale_w, length_scale, voice, f
             sid = torch.LongTensor([voice])
             audio = model_vctk.infer(x_tst, x_tst_lengths, sid=sid, noise_scale=noise_scale,
                                      noise_scale_w=noise_scale_w, length_scale=length_scale)[0][0, 0].data.cpu().float().numpy()
-            result = [ipa_text, (hps_vctk.data.sampling_rate, float2pcm(audio))]
+            result = [ipa_text, (hps_vctk.data.sampling_rate, audio)]
     if file_format == 'ogg':
         result = [result[0], wav2ogg(result[1][1], result[1][0], text, language)]
+    else:
+        result = [result[0], (result[1][0], float2pcm(result[1][1]))]
 
     return result