From 6126c23498f7d206fe944c51f66baa4a09dca981 Mon Sep 17 00:00:00 2001 From: WeberJulian Date: Mon, 20 Jun 2022 13:36:26 +0200 Subject: [PATCH] Add synpaflex formatter (#1616) * Add synpaflex formatter * Fix formatter * Make style --- TTS/tts/datasets/formatters.py | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/TTS/tts/datasets/formatters.py b/TTS/tts/datasets/formatters.py index 1e95c60f..ef05ea7c 100644 --- a/TTS/tts/datasets/formatters.py +++ b/TTS/tts/datasets/formatters.py @@ -396,6 +396,25 @@ def vctk_old(root_path, meta_files=None, wavs_path="wav48", ignored_speakers=Non return items +def synpaflex(root_path, metafiles=None, **kwargs): # pylint: disable=unused-argument + items = [] + speaker_name = "synpaflex" + root_path = os.path.join(root_path, "") + wav_files = glob(f"{root_path}**/*.wav", recursive=True) + for wav_file in wav_files: + if os.sep + "wav" + os.sep in wav_file: + txt_file = wav_file.replace("wav", "txt") + else: + txt_file = os.path.join( + os.path.dirname(wav_file), "txt", os.path.basename(wav_file).replace(".wav", ".txt") + ) + if os.path.exists(txt_file) and os.path.exists(wav_file): + with open(txt_file, "r", encoding="utf-8") as file_text: + text = file_text.readlines()[0] + items.append({"text": text, "audio_file": wav_file, "speaker_name": speaker_name}) + return items + + def open_bible(root_path, meta_files="train", ignore_digits_sentences=True, ignored_speakers=None): """ToDo: Refer the paper when available""" items = []