Uses mutagen for all audio formats

TTS/tts/datasets/dataset.py

@@ -14,6 +14,8 @@
 from TTS.utils.audio.numpy_transforms import compute_energy as calculate_energy
 
 from mutagen.mp3 import MP3
+from mutagen.flac import FLAC
+from mutagen.wave import WAVE
 
 # to prevent too many open files error as suggested here
 # https://github.com/pytorch/pytorch/issues/11201#issuecomment-421146936
@@ -48,12 +50,14 @@ def get_audio_size(audiopath):
     extension = audiopath.rpartition(".")[-1].lower()
     if extension == "mp3":
         audio_info = MP3(audiopath).info
-        return int(audio_info.length * audio_info.sample_rate)
-    if extension in ("wav", "flac"):
-        compress_factor = 8
-        bitrate = 16 # assuming 16bit audio
-        return int(os.path.getsize(audiopath) / bitrate * compress_factor)
-    raise RuntimeError(f"The audio format {extension} is not supported, please convert the audio files for mp3, flac or wav format!")
+    elif extension == "wav":
+        audio_info = WAVE(audiopath).info
+    elif extension == "flac":
+        audio_info = FLAC(audiopath).info
+    else:
+        raise RuntimeError(f"The audio format {extension} is not supported, please convert the audio files for mp3, flac or wav format!")
+
+    return int(audio_info.length * audio_info.sample_rate)
 
 
 class TTSDataset(Dataset):