File size: 4,284 Bytes
c02bdcd |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 |
import os, sys
if sys.platform == "darwin":
os.environ["PYTORCH_ENABLE_MPS_FALLBACK"] = "1"
now_dir = os.getcwd()
sys.path.append(now_dir)
from typing import Optional, List
import argparse
import numpy as np
import ChatTTS
from tools.logger import get_logger
from tools.audio import pcm_arr_to_mp3_view
from tools.normalizer.en import normalizer_en_nemo_text
from tools.normalizer.zh import normalizer_zh_tn
logger = get_logger("Command")
def save_mp3_file(wav, index):
data = pcm_arr_to_mp3_view(wav)
mp3_filename = f"output_audio_{index}.mp3"
with open(mp3_filename, "wb") as f:
f.write(data)
logger.info(f"Audio saved to {mp3_filename}")
def load_normalizer(chat: ChatTTS.Chat):
# try to load normalizer
try:
chat.normalizer.register("en", normalizer_en_nemo_text())
except ValueError as e:
logger.error(e)
except BaseException:
logger.warning("Package nemo_text_processing not found!")
logger.warning(
"Run: conda install -c conda-forge pynini=2.1.5 && pip install nemo_text_processing",
)
try:
chat.normalizer.register("zh", normalizer_zh_tn())
except ValueError as e:
logger.error(e)
except BaseException:
logger.warning("Package WeTextProcessing not found!")
logger.warning(
"Run: conda install -c conda-forge pynini=2.1.5 && pip install WeTextProcessing",
)
def main(
texts: List[str],
spk: Optional[str] = None,
stream: bool = False,
source: str = "local",
custom_path: str = "",
):
logger.info("Text input: %s", str(texts))
chat = ChatTTS.Chat(get_logger("ChatTTS"))
logger.info("Initializing ChatTTS...")
load_normalizer(chat)
is_load = False
if os.path.isdir(custom_path) and source == "custom":
is_load = chat.load(source="custom", custom_path=custom_path)
else:
is_load = chat.load(source=source)
if is_load:
logger.info("Models loaded successfully.")
else:
logger.error("Models load failed.")
sys.exit(1)
if spk is None:
spk = chat.sample_random_speaker()
logger.info("Use speaker:")
print(spk)
logger.info("Start inference.")
wavs = chat.infer(
texts,
stream,
params_infer_code=ChatTTS.Chat.InferCodeParams(
spk_emb=spk,
),
)
logger.info("Inference completed.")
# Save each generated wav file to a local file
if stream:
wavs_list = []
for index, wav in enumerate(wavs):
if stream:
for i, w in enumerate(wav):
save_mp3_file(w, (i + 1) * 1000 + index)
wavs_list.append(wav)
else:
save_mp3_file(wav, index)
if stream:
for index, wav in enumerate(np.concatenate(wavs_list, axis=1)):
save_mp3_file(wav, index)
logger.info("Audio generation successful.")
if __name__ == "__main__":
r"""
python -m examples.cmd.run \
--source custom --custom_path ../../models/2Noise/ChatTTS 你好喲 ":)"
"""
logger.info("Starting ChatTTS commandline demo...")
parser = argparse.ArgumentParser(
description="ChatTTS Command",
usage='[--spk xxx] [--stream] [--source ***] [--custom_path XXX] "Your text 1." " Your text 2."',
)
parser.add_argument(
"--spk",
help="Speaker (empty to sample a random one)",
type=Optional[str],
default=None,
)
parser.add_argument(
"--stream",
help="Use stream mode",
action="store_true",
)
parser.add_argument(
"--source",
help="source form [ huggingface(hf download), local(ckpt save to asset dir), custom(define) ]",
type=str,
default="local",
)
parser.add_argument(
"--custom_path",
help="custom defined model path(include asset ckpt dir)",
type=str,
default="",
)
parser.add_argument(
"texts",
help="Original text",
default=["YOUR TEXT HERE"],
nargs=argparse.REMAINDER,
)
args = parser.parse_args()
logger.info(args)
main(args.texts, args.spk, args.stream, args.source, args.custom_path)
logger.info("ChatTTS process finished.")
|