From 21cbfe964e31fecbfcb091e95996236455b200be Mon Sep 17 00:00:00 2001
From: Edresson Casanova <edresson1@gmail.com>
Date: Thu, 21 Apr 2022 08:26:27 -0300
Subject: [PATCH] Add RMS based norm in save_wav method

---
 TTS/server/server.py | 3 ++-
 TTS/utils/audio.py   | 6 +++++-
 2 files changed, 7 insertions(+), 2 deletions(-)

diff --git a/TTS/server/server.py b/TTS/server/server.py
index aef507fd..0f29b56b 100644
--- a/TTS/server/server.py
+++ b/TTS/server/server.py
@@ -111,7 +111,8 @@ synthesizer = Synthesizer(
     use_cuda=args.use_cuda,
 )
 
-use_multi_speaker = hasattr(synthesizer.tts_model, "num_speakers") and synthesizer.tts_model.num_speakers > 1
+use_multi_speaker = hasattr(synthesizer.tts_model, "num_speakers") and (synthesizer.tts_model.num_speakers > 1 or synthesizer.tts_speakers_file is not None)
+print("Multispeaker?", use_multi_speaker, synthesizer.tts_model.num_speakers)
 speaker_manager = getattr(synthesizer.tts_model, "speaker_manager", None)
 # TODO: set this from SpeakerManager
 use_gst = synthesizer.tts_config.get("use_gst", False)
diff --git a/TTS/utils/audio.py b/TTS/utils/audio.py
index 4d435162..fc9d1942 100644
--- a/TTS/utils/audio.py
+++ b/TTS/utils/audio.py
@@ -859,7 +859,11 @@ class AudioProcessor(object):
             path (str): Path to a output file.
             sr (int, optional): Sampling rate used for saving to the file. Defaults to None.
         """
-        wav_norm = wav * (32767 / max(0.01, np.max(np.abs(wav))))
+        if self.do_rms_norm:
+            wav_norm = self.rms_volume_norm(wav, self.db_level) * 32767
+        else:
+            wav_norm = wav * (32767 / max(0.01, np.max(np.abs(wav))))
+
         scipy.io.wavfile.write(path, sr if sr else self.sample_rate, wav_norm.astype(np.int16))
 
     def get_duration(self, filename: str) -> float: