【发布时间】：2023-02-05 23:53:55
【问题描述】：
我正在尝试在 python 中使用 Nvidia NeMo tts_en_fastpitch 模型创建 TTS。但是不能安装fastpitch模型。这些是错误：
from nemo.collections.tts.models import HifiGanModel
from nemo.collections.tts.models import FastPitchModel
spec_generator = FastPitchModel.from_pretrained("nvidia/tts_en_fastpitch")
model = HifiGanModel.from_pretrained(model_name="nvidia/tts_hifigan")
[NeMo W 2023-01-21 18:49:02 optimizers:55] Apex was not found. Using the lamb or fused_adam optimizer will error out. [NeMo W 2023-01-21 18:49:03 __init__:22] pyniniis not installed ! Please run thenemo_text_processing/setup.sh scriptprior to usage of this toolkit. Traceback (most recent call last): File "c:/Users/ASROCK B560M STEEL/Desktop/nvidia tts/ntts.py", line 3, in <module> spec_generator = FastPitchModel.from_pretrained("nvidia/tts_en_fastpitch") File "C:\Users\ASROCK B560M STEEL\AppData\Local\Programs\Python\Python38\lib\site-packages\nemo\core\classes\common.py", line 692, in from_pretrained raise FileNotFoundError(FileNotFoundError: Model nvidia/tts_en_fastpitch was not found. Check cls.list_available_models() for the list of all available models.
请帮助我。
我尝试了来自 huggingface 和 youtube 的代码。错误是一样的，tts_en_fastpitch 没有安装。
【问题讨论】：
标签： python deep-learning text-to-speech nvidia huggingface