def load_preprocess_wav(fpath_or_wav: Union[str, Path, np.ndarray]):
"""
Loads an audio file in memory and applies the same preprocessing operations used in trained
the Speaker Encoder. Using this function is not mandatory but recommended.
"""
if isinstance(fpath_or_wav, str) or isinstance(fpath_or_wav, Path):
wav = audio.load(fpath_or_wav)
else:
wav = fpath_or_wav
wav = audio.preprocess_wav(wav)
return wav
def preprocess_speaker(speaker_dir: Path):
# Give a name to the speaker that includes its dataset
speaker_name = "_".join(speaker_dir.relative_to(datasets_root).parts)
# Create an output directory with that name, as well as a txt file containing a
# reference to each source file.
speaker_out_dir = out_dir.joinpath(speaker_name)
speaker_out_dir.mkdir(exist_ok=True)
sources_fpath = speaker_out_dir.joinpath("_sources.txt")
# There's a possibility that the preprocessing was interrupted earlier, check if
# there already is a sources file.
if sources_fpath.exists():
try:
with sources_fpath.open("r") as sources_file:
existing_fnames = {
line.split(",")[0]
for line in sources_file
}
except:
existing_fnames = {}
else:
existing_fnames = {}
# Gather all audio files for that speaker recursively
sources_file = sources_fpath.open("a" if skip_existing else "w")
for in_fpath in speaker_dir.glob("**/*.%s" % extension):
# Check if the target output file already exists
out_fname = "_".join(in_fpath.relative_to(speaker_dir).parts)
out_fname = out_fname.replace(".%s" % extension, ".npy")
if skip_existing and out_fname in existing_fnames:
continue
# Load and preprocess the waveform
wav = audio.load(in_fpath)
wav = audio.preprocess_wav(wav)
if len(wav) == 0:
continue
# Create the mel spectrogram, discard those that are too short
frames = audio.wav_to_mel_filterbank(wav)
if len(frames) < partials_n_frames:
continue
out_fpath = speaker_out_dir.joinpath(out_fname)
np.save(out_fpath, frames)
logger.add_sample(duration=len(wav) / sampling_rate)
sources_file.write("%s,%s\n" % (out_fname, in_fpath))
sources_file.close()