yeyupiaoling / PaddlePaddle-DeepSpeech

基于PaddlePaddle实现的语音识别,中文语音识别。项目完善,识别效果好。支持Windows,Linux下训练和预测,支持Nvidia Jetson开发板预测。
https://yeyupiaoling.blog.csdn.net/article/details/102904306
Apache License 2.0
650 stars 143 forks source link

我下载了aishell的训练模型,并导出,但是测试打印出繁体字乱码 #97

Closed binbin901105 closed 2 years ago

binbin901105 commented 2 years ago

我下载了aishell的训练模型,并导出,但是测试打印出的是繁体字乱码,机子是没有显卡的linux机子,大神能帮忙看下么,测试的是那个压岁书的声音, image

(tensorflow33) [root@data3 PaddlePaddle-DeepSpeech-master]# python infer_path.py --wav_path=./dataset/test.wav
-----------  Configuration Arguments -----------
alpha: 1.2
beam_size: 10
beta: 0.35
cutoff_prob: 1.0
cutoff_top_n: 40
decoding_method: ctc_greedy
enable_mkldnn: False
is_long_audio: False
lang_model_path: ./lm/zh_giga.no_cna_cmn.prune01244.klm
mean_std_path: ./dataset/mean_std.npz
model_dir: ./models/infer/
to_an: True
use_gpu: False
vocab_path: ./dataset/zh_vocab.txt
wav_path: ./dataset/test.wav
------------------------------------------------
E1008 17:31:13.923094  4154 analysis_config.cc:81] Please compile with gpu to EnableGpu()
---    Fused 0 subgraphs into layer_norm op.
消耗时间:1046ms, 识别结果: 捧丫缄究丫友砷评梧砷顾钮赋荞辅咄募艳咽朽嗡算哲葬环倚塾韶倚咄加烟碳嚼磋恋钱懈淞散元顾型迅伎顾兆读鳖扇左晏震刺咄兔阔广鼠涕募午散午禄砷锭午共尔况恋锭烟懈皱烟算数觅吧评锭赡更校叠辅校何忆慢魁扇俱咄秩咕波晋貔得嚼鳖刚锭兔咄兔阔广宽玮磐兔磐兔瘤先更哨耕停琶锭缉翰涞崔缉宅午腱卧蒲催袱封淞采洱于兔阔袄币烫懈钢颗仰瞩枫绸烟采烟翁, 得分: 34
yeyupiaoling commented 2 years ago

你下载模型之后,有没有把全部的文件替换原来的文件了?

binbin901105 commented 2 years ago

image image 我刚刚把下载模型中的zh_vocab.txt mean_std.npz 又传了一遍,但执行后还是繁体字乱码,

yeyupiaoling commented 2 years ago

你删除原来的,然后重新上传吧。我怀疑是没有替换原来的。然后重新导出模型。

binbin901105 commented 2 years ago

恩,想起来了,我是先导出,再传了那两个文件,我现在重新导出下,

binbin901105 commented 2 years ago

image 大神,你很棒,现在不乱码了,