-
大佬们，第一次使用ncnn，在转换模型的时候报错
阅读源码发现部分环节会直接随机初始化输入到模型，但是模型的输入是有范围的，请问这个有什么方法可以解决吗？
## error log | 日志或报错信息 | ログ
…
-
(CDFSE) root@autodl-container-66ee44be9a-c385597c:~/CDFSE_FastSpeech2-main# python3 preprocess.py config/AISHELL3/preprocess.yaml
Processing Data ...
0%| …
-
# 🌟 New model addition
## Model description
FastSpeech2 is a TTS model that outputs mel-spectrograms given some input text. From the [paper](https://arxiv.org/abs/2006.04558) abstract:
> Non-…
-
**Describe the bug**
When training fastspeech2 with my own data, I encountered this problem in stage5:
Traceback (most recent call last):
File "/home/lhw523/anaconda3/envs/espnet/lib/python3.8/ru…
-
参考 examples/zh_en_tts/tts3 中的语音合成示例，下载了示例中的模型文件，把相关配置改成流式语音的配置项后，调用流式合成能进行部分字母及单词的合成，但有两个问题：
1、部分字母发音不准缺，如A，M，N，I，Z等
2、有非常明显的卡顿感，
这个如何解决？
conf 文件：
```sh
# This is the parameter configuration f…
-
Currently, I see this repo is an implementation of fastspeech2. Do you have a schedule to update your code with new features for singing voice synthesis?
-
There were complaints regarding the size of the m/f dataset, which was not large enough to draw conclusions. The idea then is to train a phoneme-based TTS such as [FastSpeech 2](https://github.com/coq…
-
## Describe the bug
There are a series of errors during the inference stage for FS2.
First of all, it keeps spitting out errors of not being able to locate the different json files.
Even after you …
-
1. fastspeech2推理时的batch size设置为1，这是否意味着一个请求处理结束，模型才会处理下一个请求？还是说因为async，模型能够同时对多个请求进行推理？如果是后者，与真正的batch推理在性能上是否仍有一定的差距？
2. 我也查看了其它开源的TTS项目，似乎都不支持按batch进行推理。这是否是因为TTS模型相比于LLM，在batch推理上实现比较困难？还是说batch推理…
-
Hi,
I am trying to setup Fastspeech2 with LPCNet based on this LPCTron repo. I am able to synthesise audio but audio gets clipped at the end. For LPCNet usage, custom 20dim features are required to…