from modelscope.metainfo import Trainers
from modelscope.trainers import build_trainer
from modelscope.utils.audio.audio_utils import TtsTrainType
import os

pretrained_model_id = 'demos/speech_personal_sambert-hifigan_nsf_tts_zh-cn_pretrain_16k'

dataset_id = "./temp/output_dir/"
pretrain_work_dir = "./pretrain_work_dir/"
print(os.getcwd())

训练信息，用于指定需要训练哪个或哪些模型，这里展示AM和Vocoder模型皆进行训练

目前支持训练：TtsTrainType.TRAIN_TYPE_SAMBERT, TtsTrainType.TRAIN_TYPE_VOC

训练SAMBERT会以模型最新step作为基础进行finetune

训练Vocoder（HifiGAN）会从0开始进行训练，指定多少个step，训练多少个step

train_info = {
TtsTrainType.TRAIN_TYPE_SAMBERT: { # 配置训练AM（sambert）模型
'train_steps': 202, # 训练多少个step
'save_interval_steps': 200, # 每训练多少个step保存一次checkpoint
'log_interval': 10 # 每训练多少个step打印一次训练日志
}
}

配置训练参数，指定数据集，临时工作目录和train_info

kwargs = dict(
model=pretrained_model_id, # 指定要finetune的模型
model_revision = "v1.0.4",
work_dir=pretrain_work_dir, # 指定临时工作目录
train_dataset=dataset_id, # 指定数据集id
train_type=train_info # 指定要训练类型及参数
)

trainer = build_trainer(Trainers.speech_kantts_trainer,
default_args=kwargs)

trainer.train()

HTTPError: 404 Client Error: Not Found for url: http://www.modelscope.cn/api/v1/models/demos/speech_personal_sambert-hifigan_nsf_tts_zh-cn_pretrain_16k/revisions?EndTime=1681889061