PaddleSpeech

PaddleSpeech copied to clipboard

[TTS]cudaError on Nvidia Jetson Orin NX

1

For support and discussions, please use our [Discourse forums](https://github.com/PaddlePaddle/DeepSpeech/discussions). If you've found a bug then please create an issue with the following information: **Describe the bug** When run the example...

CHNtentes

unexpected result when loading baidu_en8k

2

my code is as follows: ```python import paddle from paddlespeech.s2t.models.ds2 import DeepSpeech2Model config = './conf/deepspeech2.yaml' model = DeepSpeech2Model.from_config(config) model_param_path = './baidu_en8k/params.pdparams' model_dict = paddle.load(model_param_path) print(model_dict) model.set_state_dict(model_dict) ``` The error when...

firdota

【asr】处理使用 PaddleSpeech 英文语音识别中出现的报错 ValueError (InvalidArgument) Broadcast dimension mismatch

2

官网示例中都是中文语音识别 ``` >>> from paddlespeech.cli.asr.infer import ASRExecutor >>> asr = ASRExecutor() >>> result = asr(audio_file="zh.wav") >>> print(result) 我认为跑步最重要的就是给我带来了身体健康 ``` 当然用这份代码跑中文也会报错，需要result = asr(audio_file="zh.wav")加上指定的model模型 这里感谢提出bug和修复的#3697 但是在应用到英文语音识别是会发现找不到模型，将infer.py文件中所有的“zh”替换成“en”也没用，在生成tag时还是会变成zh后缀 因此需要将/PaddleSpeech/paddlespeech/cli/asr/infer.py文件159行 ``` # tag = model_type...

chenzf11

【问题解决】解决PaddleSpeech历史问题和BUG

2

### 背景 paddlespeech目前存在一些历史问题以及长期issue，对这些问题和issue进行修复。 ### 现状 1. 对2023.4.10~2024.4.10的issue进行统计和分类，对其中**未解决的BUG**进行处理。 2. 基于paddlepaddle-gpu\=\=2.6.1 paddlespeech\=\=develop对demos进行验证，对发现的问题进行修复。 ### issue列表 | 序号 | issue | 说明 | 认领人/状态/PR号 | | ---- | -------------------------------------------------------- | -------- | ---------------- |...

kk-2000

Bump axios from 0.26.1 to 0.28.0 in /demos/speech_web/web_client

1

Bumps [axios](https://github.com/axios/axios) from 0.26.1 to 0.28.0. Release notes Sourced from axios's releases. Release v0.28.0 Release notes: Bug Fixes fix(security): fixed CVE-2023-45857 by backporting withXSRFToken option to v0.x (#6091) Backports from...

dependabot[bot]

Bump vite from 2.9.1 to 2.9.18 in /demos/speech_web/web_client

1

Bumps [vite](https://github.com/vitejs/vite/tree/HEAD/packages/vite) from 2.9.1 to 2.9.18. Changelog Sourced from vite's changelog. 2.9.18 (2024-03-24) fix: port #15653 to v2 (#15657) (1f855dc), closes #15653 #15657 fix: port #16250 to v2 (#16254) (011bbca),...

dependabot[bot]

Bump follow-redirects from 1.14.9 to 1.15.6 in /demos/speech_web/web_client

1

Bumps [follow-redirects](https://github.com/follow-redirects/follow-redirects) from 1.14.9 to 1.15.6. Commits 35a517c Release version 1.15.6 of the npm package. c4f847f Drop Proxy-Authorization across hosts. 8526b4a Use GitHub for disclosure. b1677ce Release version 1.15.5 of...

dependabot[bot]

[S2T] paddlespeech 1.4.1 在Paddle 2.5.1环境中 示例跑不通

9

换了三台机器，捣鼓了一天时间 环境： Windows10 / 11 Python 3.10 Paddle 2.5.1 GPU版 （conda install paddlepaddle-gpu==2.5.1 cudatoolkit=11.7 -c https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/Paddle/ -c conda-forge） paddlespeech 1.4.1 （pip install paddlespeech） 问题： 1 运行 paddlespeech asr --input zh.wav，先是抛出了一个numpy异常：module...

geniuszxd

[TTS]微调报错现存不足

3

   微调训练语料1000条 batch_size是16 epoch是30 请问下1000条语料内存占用大概多大呢，是基于fastspeech2_mix权重的微调

wytyl13

给定了gpu的参数，但是gpu没得到利用，并且batch_size为128，12小时训练了iter 800/6420，这是常规的速度吗？

abcdbosh