Medix-AI/coqui-tts
Medix-AI
/
coqui-tts
mirror of https://github.com/coqui-ai/TTS.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Update VCTK formatter

This commit is contained in:
Eren Gölge 2021-12-08 14:45:32 +00:00
parent 420b92d5ae
commit 28d98da422
1 changed files with 23 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

37
TTS/tts/datasets/formatters.py
View File

@ -289,8 +289,10 @@ def brspeech(root_path, meta_file, ignored_speakers=None):
return items return items
def vctk(root_path, meta_files=None, wavs_path="wav48", ignored_speakers=None): def vctk(root_path, meta_files=None, wavs_path="wav22", mic="mic2", ignored_speakers=None):
"""homepages.inf.ed.ac.uk/jyamagis/release/VCTK-Corpus.tar.gz""" """https://datashare.ed.ac.uk/bitstream/handle/10283/3443/VCTK-Corpus-0.92.zip"""
file_ext = 'flac'
test_speakers = meta_files
items = [] items = []
meta_files = glob(f"{os.path.join(root_path,'txt')}/**/*.txt", recursive=True) meta_files = glob(f"{os.path.join(root_path,'txt')}/**/*.txt", recursive=True)
for meta_file in meta_files: for meta_file in meta_files:
@ -302,26 +304,33 @@ def vctk(root_path, meta_files=None, wavs_path="wav48", ignored_speakers=None):
continue continue
with open(meta_file, "r", encoding="utf-8") as file_text: with open(meta_file, "r", encoding="utf-8") as file_text:
text = file_text.readlines()[0] text = file_text.readlines()[0]
wav_file = os.path.join(root_path, wavs_path, speaker_id, file_id + ".wav") # p280 has no mic2 recordings
items.append({"text": text, "audio_file": wav_file, "speaker_name": "VCTK_" + speaker_id}) if speaker_id == "p280":
wav_file = os.path.join(root_path, wavs_path, speaker_id, file_id + f"_mic1.{file_ext}")
else:
wav_file = os.path.join(root_path, wavs_path, speaker_id, file_id + f"_{mic}.{file_ext}")
if os.path.exists(wav_file):
items.append([text, wav_file, "VCTK_" + speaker_id])
else:
print(f" [!] wav files don't exist - {wav_file}")
return items return items
def vctk_slim(root_path, meta_files=None, wavs_path="wav48", ignored_speakers=None): # pylint: disable=unused-argument def vctk_old(root_path, meta_files=None, wavs_path="wav48"):
"""homepages.inf.ed.ac.uk/jyamagis/release/VCTK-Corpus.tar.gz""" """homepages.inf.ed.ac.uk/jyamagis/release/VCTK-Corpus.tar.gz"""
test_speakers = meta_files
items = [] items = []
txt_files = glob(f"{os.path.join(root_path,'txt')}/**/*.txt", recursive=True) meta_files = glob(f"{os.path.join(root_path,'txt')}/**/*.txt", recursive=True)
for text_file in txt_files: for meta_file in meta_files:
_, speaker_id, txt_file = os.path.relpath(text_file, root_path).split(os.sep) _, speaker_id, txt_file = os.path.relpath(meta_file, root_path).split(os.sep)
file_id = txt_file.split(".")[0] file_id = txt_file.split(".")[0]
# ignore speakers if isinstance(test_speakers, list): # if is list ignore this speakers ids
if isinstance(ignored_speakers, list): if speaker_id in test_speakers:
if speaker_id in ignored_speakers:
continue continue
with open(meta_file, "r", encoding="utf-8") as file_text:
text = file_text.readlines()[0]
wav_file = os.path.join(root_path, wavs_path, speaker_id, file_id + ".wav") wav_file = os.path.join(root_path, wavs_path, speaker_id, file_id + ".wav")
items.append([None, wav_file, "VCTK_" + speaker_id]) items.append([text, wav_file, "VCTK_old_" + speaker_id])
return items return items