AI-应用大全

图片中文识别
https://github.com/PaddlePaddle/PaddleOCR
https://github.com/breezedeus/cnocr
python3 scripts/cnocr_predict.py --file text.png
文字处理
https://github.com/hankcs/HanLP
https://github.com/hankcs/pyhanlp
https://github.com/ownthink/Jiagu
分词
hanlp segment <<< ‘欢迎新老师生前来就餐’
句法分析
hanlp parse <<< ‘欢迎新老师生前来就餐’
关键词提取
HanLP.extractKeyword(‘欢迎新老师生前来就餐’, 2)
自动摘要
HanLP.extractSummary(‘欢迎新老师生前来就餐’, 3)
依存句法分析
HanLP.parseDependency(‘欢迎新老师生前来就餐’)
语音识别
https://github.com/nl8590687/ASRT_SpeechRecognition
python3 asrserver.py
https://github.com/kaldi-asr/kaldi
http://kaldi-asr.org/
https://github.com/jackyyy0228/Chinese-ASR
TTS
https://github.com/espnet/espnet

预训练模型镜像
https://coggle.club/note/dl/pretrained-models

文字转语音
大声说一句：
say { {“I like to ride my bike.”}}
大声朗读文件：
say -f { {filename.txt}}
播放自定义语音和语速的短语：
say -v { {voice}} -r { {words_per_minute}} { {“I’m sorry Dave, I can’t let you do that.”}}
列出可用的声音：
say -v ?
创建语音文本的音频文件：
say -o { {filename.aiff}} { {“Here’s to the Crazy Ones.”}}

清华大学自然语言处理
https://github.com/thunlp/THULAC-Python
python3 test.py
python3 -m thulac input.txt output.txt
结巴分词
https://github.com/fxsjy/jieba
处理中文文本内容
https://github.com/isnowfy/snownlp
https://github.com/tsroten/pynlpir

https://github.com/stacklikemind/deepnude_official
https://github.com/lwlodo/deep_nude/
https://github.com/emperorwushi/xi/
https://github.com/NVIDIA/pix2pixHD

Text-To-Speech

多语言：
https://github.com/pndurette/gTTS
pip install gTTS
gtts-cli ‘hello’ --output hello.mp3
https://github.com/cboard-org/cboard

https://github.com/zlargon/google-tts/
npm install google-tts-api --save
https://github.com/vilic/cordova-plugin-tts
https://github.com/naoufal/react-native-speech
web:
https://github.com/guest271314/SpeechSynthesisRecorder
https://github.com/kripken/speak.js
https://github.com/Marak/say.js
命令行：
https://www.npmjs.com/package/voc-cli
py:
https://github.com/buriburisuri/speech-to-text-wavenet
https://github.com/readbeyond/aeneas
http://espeak.sourceforge.net/test/latest.html
https://github.com/Kyubyong/tacotron
完全端到端的文本到语音合成模型，主要是将文本转化为语音，使用了预训练模型(pre-trained)技术
https://github.com/keithito/tacotron
https://keithito.com/LJ-Speech-Dataset/
https://librivox.org/
https://github.com/DragonComputer/Dragonfire

https://github.com/r9y9/deepvoice3_pytorch
git clone https://github.com/r9y9/deepvoice3_pytorch && cd deepvoice3_pytorch
pip install -e “.[bin]”
python synthesis.py --preset=20180505_deepvoice3_ljspeech.json
20180505_deepvoice3_checkpoint_step000640000.pth
sentences.txt
output_dir
python preprocess.py --preset=presets/deepvoice3_ljspeech.json ljspeech ~/data/LJSpeech-1.0
python train.py --preset=presets/deepvoice3_ljspeech.json --data-root=./data/ljspeech
python preprocess.py ljspeech ~/data/LJSpeech-1.0

warning! this may use different hyper parameters used at preprocessing stage

python train.py --preset=presets/deepvoice3_ljspeech.json --data-root=./data/ljspeech
https://github.com/mozilla/TTS
https://github.com/hgneng/ekho #Chinese
http://www.eguidedog.net/ekho.php

speech-to-text

https://github.com/mozilla/DeepSpeech
pip3 install deepspeech
deepspeech --model models/output_graph.pbmm --alphabet models/alphabet.txt --lm models/lm.binary --trie models/trie --audio my_audio_file.wav
pip3 install deepspeech-gpu
deepspeech --model models/output_graph.pbmm --alphabet models/alphabet.txt --lm models/lm.binary --trie models/trie --audio my_audio_file.wav
pre-trained model
wget https://github.com/mozilla/DeepSpeech/releases/download/v0.1.1/deepspeech-0.1.1-models.tar.gz
tar xvfz deepspeech-0.1.1-models.tar.gz
audio files
wget https://github.com/mozilla/DeepSpeech/releases/download/v0.1.1/audio-0.1.1.tar.gz
tar xvfz audio-0.1.1.tar.gz
deepspeech models/output_graph.pb audio/2830-3980-0043.wav models/alphabet.txt models/lm.binary models/trie
Text: experience proves this
deepspeech models/output_graph.pb audio/4507-16021-0012.wav models/alphabet.txt models/lm.binary models/trie
Text: why should one halt on the way
deepspeech models/output_graph.pb audio/8455-210777-0068.wav models/alphabet.txt models/lm.binary models/trie
Text: your power is sufficient i said
deepspeech --model models/output_graph.pbmm --alphabet models/alphabet.txt --lm models/lm.binary --trie models/trie --audio my_audio_file.wav
To download the pre-built binaries, use util/taskcluster.py:
python3 util/taskcluster.py --target .
or if you’re on macOS:
python3 util/taskcluster.py --arch osx --target .

https://github.com/asticode/go-astideepspeech
http://www.cstr.ed.ac.uk/projects/festival/

采用densenet识别图中文字
https://github.com/yinchangchang/ocr_densenet

阿里云语音验证码
https://github.com/qingdie/qingdie-aliyun
https://dysmsapi.aliyuncs.com/ #短信验证码
https://dyvmsapi.aliyuncs.com/ #语音验证码

https://github.com/kerlomz/captcha_trainer
https://github.com/kerlomz/captcha_library_c
https://github.com/kerlomz/captcha_demo_csharp
https://github.com/kerlomz/captcha_platform
https://mp.weixin.qq.com/s/6IAEus9OTg-hP9NGKJRm_Q

字典大全
http://www.zd9999.com/
https://github.com/GopherCoder/dictionary-of-chinese
https://github.com/pwxcoo/chinese-xinhua

古诗词
https://github.com/chinese-poetry/chinese-poetry
https://github.com/KomaBeyond/chinese-poetry-mysql
https://github.com/Werneror/Poetry

泼辣有图
http://www.polayoutu.com/collections

查天气
https://github.com/tangjiahao/robotofwx/blob/master/robotmain.py
https://api.seniverse.com/v3/weather/now.json?key=Skb40T46PiBDM35V2&location=%s&language=zh-Hans&unit=c
https://free-api.heweather.net/s6/weather?location=%s&key=a3269a0918a44a62ae97c314dd24f02a

酷狗音乐
http://www.kugou.com/yy/index.php?r=play/getdata&hash=%s&album_id=%s&=1497972864535
https://wwwapi.kugou.com/yy/index.php?r=play/getdata&callback=jQuery191014887140948582345_1557824383110&hash=%s&album_id=%s&dfid=0zpwSa44LtGp0D89Gr371MJb&mid=51eafc9b0e5eaca4e106b905175401ec&platid=4&=1557824383112
http://songsearch.kugou.com/song_search_v2?keyword=%spage=1&pagesize=3&userid=-1&clientver=&platform=WebFilter&tag=em&filter=2&iscorrection=1&privilege_filter=0
https://github.com/tangjiahao/robotofwx/blob/master/robotmain.py

机器学习
https://github.com/eriklindernoren/ML-From-Scratch
https://github.com/NELSONZHAO/zhihu

机器翻译machine translation（NMT）
https://github.com/tensorflow/nmt
https://github.com/OpenNMT/OpenNMT-py
http://opennmt.net/OpenNMT-py/speech2text.html
https://github.com/OpenNMT/OpenNMT
https://github.com/THUNLP-MT/MT-Reading-List
https://github.com/xuwenshen/Machine-Translation
https://github.com/foamliu/Machine-Translation-v2

英汉词典
https://github.com/ChestnutHeng/Wudao-dict
https://github.com/program-in-chinese/vscode_english_chinese_dictionary
https://github.com/skywind3000/ECDICT
https://github.com/skywind3000/ECDICT/releases
https://github.com/program-in-chinese/webextension_english_chinese_dictionary
https://github.com/fxsjy/diaosi
https://github.com/chienlungcheung/MyDict

手机短信验证码语音验证码话费充值流量充值
https://github.com/gitchenze/panguPhone
http://www.miaodiyun.com/

语音识别
https://github.com/xxbb1234021/speech_recognition
训练数据下载 清华大学中文语料库（thchs30）http://www.openslr.org/18/
训练
配置conf目录下的conf.ini文件中的各项
在终端运行 python train.py 开始训练
在终端运行 python test.py 测试
也可以使用PyCharm打开

wav 文件转 16k 16bits 位深的单声道pcm文件
ffmpeg -y -i 16k.wav -acodec pcm_s16le -f s16le -ac 1 -ar 16000 16k.pcm
44100 采样率 单声道 16bts pcm 文件转 16000采样率 16bits 位深的单声道pcm文件
ffmpeg -y -f s16le -ac 1 -ar 44100 -i test44.pcm -acodec pcm_s16le -f s16le -ac 1 -ar 16000 16k.pcm
mp3 文件转 16K 16bits 位深的单声道 pcm文件
ffmpeg -y -i aidemo.mp3 -acodec pcm_s16le -f s16le -ac 1 -ar 16000 16k.pcm
// -acodec pcm_s16le pcm_s16le 16bits 编码器 // -f s16le 保存为16bits pcm格式 // -ac 1 单声道 // -ar 16000 16000采样率
Facebook AI Research的自动语音识别工具包
https://github.com/facebookresearch/wav2letter
https://github.com/brightmart/roberta_zh

https://github.com/facebookresearch/SlowFast 视频分类/视频理解/行为检测

中文语音识别 AISHELL
https://github.com/libai3/masr
识别自己的语音
brew install portaudio
pip3 install pyaudio
语言模型
https://deepspeech.bj.bcebos.com/zh_lm/zh_giga.no_cna_cmn.prune01244.klm

https://github.com/Uberi/speech_recognition
https://pypi.org/project/pocketsphinx/
pip3 install SpeechRecognition
pip3 install https://github.com/bambocher/pocketsphinx-python/archive/master.zip
brew install cmu-pocketsphinx cmu-sphinxbase cmu-sphinxtrain cmuclmtk
https://realpython.com/python-speech-recognition/
https://blog.csdn.net/weixin_40490238/article/details/84841825
https://sourceforge.net/projects/cmusphinx/files/Acoustic%20and%20Language%20Models/Mandarin/cmusphinx-zh-cn-5.2.tar.gz/download
https://jaist.dl.sourceforge.net/project/cmusphinx/Acoustic%20and%20Language%20Models/Mandarin/cmusphinx-zh-cn-5.2.tar.gz
解压到
cd /usr/local/lib/python3.7/site-packages/speech_recognition/pocketsphinx-data
mkdir -p zh-CN/acoustic-model
zh_broadcastnews_16k_ptm256_8000.tar.bz2解压缩到zh-CN/acoustic-model
zh_broadcastnews_utf8.dic重命名为pronounciation-dictionary.dict并放入\zh-CN文件夹
SphinxBase工具将zh_broadcastnews_64000_utf8.DMP转换成language-model.lm.bin并放入\zh-CN文件夹下
pocketsphinx_continuous -hmm /usr/local/lib/python3.7/site-packages/speech_recognition/pocketsphinx-data/zh-CN/acoustic-model/ -lm zh_broadcastnews_64000_utf8.DMP -dict pronounciation-dictionary.dic
pocketsphinx_continuous -hmm zh_broadcastnews_ptm256_8000 -lm zh_broadcastnews_64000_utf8.DMP -dict zh_broadcastnews_utf8.dic -infile myfile-16000.wav > myfile.txt
pocketsphinx_continuous -inmic yes -hmm …/share/pocketsphinx/model/cmusphinx-zh-cn-5.2/zh_cn.cd_cont_5000 -lm …/share/pocketsphinx/model/cmusphinx-zh-cn-5.2/zh_cn.lm.bin -dict …/share/pocketsphinx/model/cmusphinx-zh-cn-5.2/zh_cn.dic
pocketsphinx_continuous -inmic yes -hmm /usr/local/pocketsphinx/share/pocketsphinx/model/cmusphinx-zh-cn-5.2/zh_cn.cd_cont_5000 -lm ./4648.lm -dict ./4648.dic

https://github.com/cmusphinx/sphinxbase
.\sphinx_lm_convert.exe -i .\zh_broadcastnews_64000_utf8.DMP -o language-model.lm -ofmt arpa
.\sphinx_lm_convert.exe -i .\language-model.lm -o language-model.lm.bin
sphinx_lm_convert -i zh_broadcastnews_64000_utf8.DMP -o language-model.lm -ofmt arpa
sphinx_lm_convert -i language-model.lm -o language-model.lm.bin
https://www.cnblogs.com/henjay724/p/9576670.html
http://www.speech.cs.cmu.edu/tools/lextool.html
http://www.speech.cs.cmu.edu/tools/lmtool-new.html
pip3 install cmudict
vi test.txt
窗口 ch uang k ou
打开 d a k ai
关闭 g uan b i
记事本 j i sh ib b en
浏览器 l iu l an q i
音乐 y in uxs uxe
http://www.speech.cs.cmu.edu/tools/lmtool-new.html
pocketsphinx_continuous -lm 6177.lm -dict 6177.dic
/usr/local/Cellar/cmu-pocketsphinx/0.8/share/pocketsphinx/model/hmm/en_US/hub4wsj_sc_8k
pocketsphinx_continuous -hmm tdt_sc_8k -lm 6177.lm -dict 6177.dic
/usr/local/share/pocketsphinx/model/hmm/zh/tdt_sc_8k/
/usr/local/Cellar/cmu-pocketsphinx/0.8/share/pocketsphinx/model/hmm/zh/tdt_sc_8k/
训练大文本数据的语言模型
vi weather.txt
天气
有雨
晴朗
多云
雷电
产生词汇表vocabulary文件：
text2wfreq < weather.txt | wfreq2vocab > weather.vocab
命令text2wfreq：统计文本文件中每个词出现的次数，得到一个后缀为wfreq的文件
命令wfreq2vocab：统计文本文件中含有多少个词，即有哪些词。
生成 arpa格式的语言模型：
text2idngram -vocab weather.vocab -idngram weather.idngram < weather.txt
idngram2lm -vocab_type 0 -idngram weather.idngram -vocab weather.vocab -arpa weather.arpa
转换为 CMU的二进制格式 (DMP)：
sphinx_lm_convert -i weather.arpa -o weather.lm.DMP
cp -a /usr/local/share/pocketsphinx/model/hmm/zh/tdt_sc_8k .
sphinx_fe -argfile tdt_sc_8k/feat.params -samprate 16000 -c arctic20.fileids -di . -do . -ei wav -eo mfc -mswav yes
https://www.cnblogs.com/qiuhong/articles/3671991.html
sphinx_lm_convert -i model.lm -o model.dmp
sphinx_lm_convert -i model.dmp -ifmt dmp -o model.lm -ofmt arpa
http://www.voidcn.com/article/p-tiryhtrm-zk.html
rec_wav.sh
for i in seq 1 12; do
fn=printf arctic_%04d $i;
read sent; echo $sent;
rec -r 16000 -e signed-integer -b 16 -c 1 $fn.wav 2>/dev/null;
done < arctic20.txt

Raspberry PI语音控制-PocketSphinx
https://my.oschina.net/RagingTyphoon/blog/493072

IBM:
https://github.com/watson-developer-cloud/speech-to-text-nodejs
https://stream-wdc.watsonplatform.net/speech-to-text/api
https://gateway-syd.watsonplatform.net/speech-to-text/api
https://speech-to-text-demo.ng.bluemix.net/

GOOGLE:
https://console.developers.google.com/
http://www.chromium.org/developers/how-tos/api-keys
https://github.com/gillesdemey/google-speech-v2
brew install sox
rec --encoding signed-integer --bits 16 --channels 1 --rate 16000 test.wav
curl -X POST
–data-binary @‘audio/hello (16bit PCM).wav’
–header ‘Content-Type: audio/l16; rate=16000;’
‘https://www.google.com/speech-api/v2/recognize?output=json&lang=en-us&key=yourkey’
curl -X POST
–data-binary @audio/good-morning-google.flac
–header ‘Content-Type: audio/x-flac; rate=44100;’
‘https://www.google.com/speech-api/v2/recognize?output=json&lang=en-us&key=yourkey’
https://github.com/evancohen/sonus
npm install --save sonus

科大讯飞
https://www.xfyun.cn/
https://www.xfyun.cn/services/voicedictation
http://member.voicecloud.cn/index.php/default/register
https://www.xfyun.cn/solutions/robots
http://www.devstore.cn/evaluation/testInfo/107-127.html

文字转拼音
https://github.com/janx/ruby-pinyin
https://github.com/sofish/han

微信CLIENT
https://github.com/trazyn/weweChat #只有PC
开源IM
https://github.com/hcxiong/xuanxuan #只有PC
https://github.com/meili/TeamTalk
https://github.com/YiChat
https://github.com/duckchat/gaga
https://github.com/dianbaer/anychat
https://github.com/zulip

https://github.com/gunthercox/ChatterBot
https://github.com/pandolia/qqbot
https://github.com/huangzk/qqchatbot
https://gitee.com/airgzn/QQChatBot
https://gitee.com/airgzn/xiaofeichatbot

APPLE
https://developer.apple.com/documentation/avfoundation/speech_synthesis
https://github.com/CoderTitan/TextAndVoice

http://ai.youdao.com/
https://openapi.youdao.com/api
https://openapi.youdao.com/ocrtransapi
https://openapi.youdao.com/speechtransapi
https://openapi.youdao.com/ocrapi
https://openapi.youdao.com/ocr_structure
https://openapi.youdao.com/ocr_formula
https://www.cnblogs.com/alchemystar/p/13668470.html
http://ai.youdao.com/DOCSIRMA/html/%E8%87%AA%E7%84%B6%E8%AF%AD%E8%A8%80%E7%BF%BB%E8%AF%91/API%E6%96%87%E6%A1%A3/%E6%96%87%E6%9C%AC%E7%BF%BB%E8%AF%91%E6%9C%8D%E5%8A%A1/%E6%96%87%E6%9C%AC%E7%BF%BB%E8%AF%91%E6%9C%8D%E5%8A%A1-API%E6%96%87%E6%A1%A3.html
http://ai.youdao.com/DOCSIRMA/html/%E8%87%AA%E7%84%B6%E8%AF%AD%E8%A8%80%E7%BF%BB%E8%AF%91/API%E6%96%87%E6%A1%A3/%E8%AF%AD%E9%9F%B3%E7%BF%BB%E8%AF%91%E6%9C%8D%E5%8A%A1/%E8%AF%AD%E9%9F%B3%E7%BF%BB%E8%AF%91%E6%9C%8D%E5%8A%A1-API%E6%96%87%E6%A1%A3.html
http://ai.youdao.com/DOCSIRMA/html/%E8%87%AA%E7%84%B6%E8%AF%AD%E8%A8%80%E7%BF%BB%E8%AF%91/API%E6%96%87%E6%A1%A3/%E5%9B%BE%E7%89%87%E7%BF%BB%E8%AF%91%E6%9C%8D%E5%8A%A1/%E5%9B%BE%E7%89%87%E7%BF%BB%E8%AF%91%E6%9C%8D%E5%8A%A1-API%E6%96%87%E6%A1%A3.html
http://ai.youdao.com/DOCSIRMA/html/%E6%96%87%E5%AD%97%E8%AF%86%E5%88%ABOCR/API%E6%96%87%E6%A1%A3/%E9%80%9A%E7%94%A8OCR%E6%9C%8D%E5%8A%A1/%E9%80%9A%E7%94%A8OCR%E6%9C%8D%E5%8A%A1-API%E6%96%87%E6%A1%A3.html

BAIDU

https://github.com/Baidu-AIP/nodejs-sdk
npm install baidu-aip-sdk
http://ai.baidu.com/docs#/
http://yuyin.baidu.com/
https://github.com/ChenHao96/VoiceInteraction
https://github.com/eisneim/cytron_tts_gui
https://github.com/apetab/vbot-voice

http://tsn.baidu.com/text2audio
QQ&微信语音silk转换wav
brew install gcc ffmpeg
git clone https://github.com/kn007/silk-v3-decoder.git silk-v3-decoder
cd silk-v3-decoder/silk
make && make decoder
./decoder 123.silk 123.pcm
ffmpeg -y -f s16le -ar 24000 -ac 1 -i 123.pcm -f wav -ar 16000 -b:a 16 -ac 1 123.wav
https://www.jianshu.com/p/b092da81feb0
语音识别
len + speech方式
http://vop.baidu.com/server_api?format=wav&rate=16000&channel=1&token=&cuid=9e:eb:e8:d4:67:00&len=大小&speech=图片base64
url + callback方式
http://vop.baidu.com/server_api?format=wav&rate=16000&channel=1&token=&cuid=9e:eb:e8:d4:67:00&url=123.wav&callback=回调地址
http://tts.baidu.com/text2audio?lan=zh&ie=UTF-8&spd=2&text=
https://ai.baidu.com/aidemo?type=tns2&idx=1&tex=%s&cuid=baidu_speech_demo&cod=2&lan=zh&ctp=1&pdt=1&spd=5&per=4&vol=5&pit=5
http://tts.baidu.com/text2audio?lan=zh&ie=UTF-8&spd=2&text=你要转换的文字
https://openapi.baidu.com/oauth/2.0/token?grant_type=client_credentials&client_id={}&client_secret={}&

https://github.com/nl8590687/ASRT_SpeechRecognition
cp -rf datalist/* dataset/
目前可用的模型有24、25和251
本项目开始训练请执行：
$ python3 train_mspeech.py
本项目开始测试请执行：
$ python3 test_mspeech.py
测试之前，请确保代码中填写的模型文件路径存在。
ASRT API服务器启动请执行：
$ python3 asrserver.py
如果要训练和使用模型251，请在代码中 import SpeechModel 的相应位置做修改。
dataset/data_thchs30/train/.wav
dataset/data_thchs30/dev/.wav
dataset/data_thchs30/test/.wav
dataset/ST-CMDS-20170001_1-OS/.wav
https://github.com/nl8590687/ASRT_SpeechRecognition/wiki

https://github.com/apachecn/AiLearning
https://feisky.xyz/machine-learning/
自然语言处理 中文分词 词性标注 命名实体识别 依存句法分析 关键词提取 新词发现 短语提取 自动摘要 文本分类 拼音简繁
http://hanlp.com/
https://github.com/hankcs/HanLP
https://github.com/hankcs/pyhanlp
pip3 install pyhanlp
hanlp update
hanlp --help
hanlp segment <<< ‘欢迎新老师生前来就餐’
hanlp parse <<< ‘徐先生还具体帮助他确定了把画雄鹰、松鼠和麻雀作为主攻目标。’
https://github.com/fighting41love/cocoNLP

https://github.com/ownthink/Jiagu

https://github.com/PaddlePaddle/LARK/tree/develop/ERNIE
pip3 install paddlepaddle
https://mp.weixin.qq.com/s/nb3g1RV3fk2rm8a_v_ZUEA
https://mp.weixin.qq.com/s/osfV54FRU1vw5c4CZuSR1A
https://github.com/PaddlePaddle/book
http://www.paddlepaddle.org/documentation/docs/zh/1.2/beginners_guide/quick_start/index.html

https://github.com/explosion/spaCy

https://github.com/visipedia/iwildcam_comp
https://github.com/visipedia/inat_comp
https://github.com/macaodha/inat_comp_2018
https://github.com/Microsoft/AirSim
https://www.microsoft.com/en-us/ai/ai-for-earth?activetab=pivot1%3aprimaryr6
http://cocodataset.org/#download
https://hackaday.io/project/159737-spectra-open-biomedical-imaging

人工智能开发平台
https://github.com/ifeegoo/Prometheus
https://github.com/huanghe/ai
https://mp.weixin.qq.com/s/-y_01EBYVxiCwLddCvyFfg
https://github.com/intel-analytics/analytics-zoo
https://analytics-zoo.github.io/0.4.0/

以太坊智能合约+DApp 工作流实战案例：抽奖程序
https://github.com/wangshijun/ethereum-lottery-dapp
https://infura.io/project/08ed39a60be74cd78974ecfed000fe6f
https://infura.io/docs/gettingStarted/authentication
npm install wscat -g
wscat -c wss://mainnet.infura.io/ws/v3/08ed39a60be74cd78974ecfed000ff

{“jsonrpc”: “2.0”, “id”: 1, “method”: “eth_blockNumber”, “params”: []}

npm install -g solc truffle ganache-cli
https://github.com/trufflesuite/ganache-cli
https://truffleframework.com/ganache
https://github.com/wangzukun/truffle4-demo
truffle init

推荐
https://mp.weixin.qq.com/s/E6EH6aJjzTwN2UZf_4nwoA

Synonyms 中文近义词工具包，可以用于自然语言理解的很多任务：文本对齐，推荐算法，相似度计算，语义偏移，关键字提取，概念提取，自动摘要，搜索引擎等
https://github.com/huyingxi/Synonyms

语言/知识表示工具
https://github.com/PaddlePaddle/LARK

句子、QA相似度匹配
https://github.com/NTMC-Community/MatchZoo

https://polyglot.readthedocs.io/en/latest/Installation.html
https://github.com/aboSamoor/polyglot
brew install polyglot
pip3 install polyglot

pyltp
https://github.com/HIT-SCIR/pyltp
https://mp.weixin.qq.com/s/gLzdYZVoegjAPmnMAUq19g
pip3 install pyltp
https://pyltp.readthedocs.io/zh_CN/develop/api.html

反向传递: https://www.cnblogs.com/charlotte77/p/5629865.html
CNN原理: http://www.cnblogs.com/charlotte77/p/7759802.html
RNN原理: https://blog.csdn.net/qq_39422642/article/details/78676567
LSTM深入浅出的好文: https://blog.csdn.net/roslei/article/details/61912618

语音翻译 面对面翻译小程序
https://github.com/Tencent/Face2FaceTranslator

中英文翻译
https://github.com/xuwenshen/Machine-Translation
https://github.com/liuhuanyong/ChineseTextualInference
https://github.com/quincyliang/nlp-public-dataset

中文语料
https://github.com/yanwii/machine-translation
https://github.com/brightmart/nlp_chinese_corpus

https://github.com/FeeiCN/dict

http://www.iciba.com/
http://www.iciba.com/hello
https://github.com/justinyhuang/BashCiba

https://github.com/Neoyyy/google-CommandLine-Translation-Tool
https://translate.google.cn/translate_a/single?hl=zh-CN&sl=zh-CN&tl=en&q=%E4%B8%AD%E5%9B%BD&client=tw-ob

http://www.baidu.com/
http://fanyi.baidu.com/basetrans
https://github.com/AnuoF/TranslateTool

http://fanyi.youdao.com/openapi?path=data-mode
http://fanyi.youdao.com/openapi.do?keyfrom=wufeifei&key=716426270&type=data&doctype=json&version=1.1&q=测试

代码安全审计
https://github.com/WhaleShark-Team/cobra

XLNET/NLP预训练新方法
自编码语言模型（Autoencoder LM）
https://github.com/zihangdai/xlnet

bert中文分类实践/ELMO
自回归语言模型（Autoregressive LM）
https://github.com/NLPScott/bert-Chinese-classification-task
https://github.com/yuanxiaosc/BERT_Paper_Chinese_Translation
https://github.com/terrifyzhao/bert-utils
http://icrc.hitsz.edu.cn/info/1037/1162.htm
https://github.com/NVIDIA/Megatron-LM

https://github.com/ymcui/Chinese-BERT-wwm
以TensorFlow版本为例，下载完毕后对zip文件进行解压得到：
chinese_wwm_L-12_H-768_A-12.zip
|- bert_model.ckpt # 模型权重
|- bert_model.meta # 模型meta信息
|- bert_model.index # 模型index信息
|- bert_config.json # 模型参数
|- vocab.txt # 词表
https://github.com/ymcui/cmrc2018
CMRC 2018数据集是哈工大讯飞联合实验室发布的中文机器阅读理解数据。根据给定问题，系统需要从篇章中抽取出片段作为答案，形式与SQuAD相同。
https://github.com/DRCKnowledgeTeam/DRCD
DRCD数据集由中国台湾台达研究院发布，其形式与SQuAD相同，是基于繁体中文的抽取式阅读理解数据集。
https://github.com/shiyybua/NER
中文命名实体识别（NER）任务中，我们采用了经典的人民日报数据以及微软亚洲研究院发布的NER数据。
THUCNews
http://thuctc.thunlp.org/
由清华大学自然语言处理实验室发布的新闻数据集，需要将新闻分成10个类别中的一个。

识别/塑造面部
https://deepfakes.com.cn/
https://deepfakes.com.cn/index.php/95.html
https://deepfakes.com.cn/index.php/265.html
https://deepfakes.com.cn/index.php/243.html
https://github.com/deepfakes/faceswap
https://github.com/deepfakes/faceswap/blob/master/INSTALL.md
从您的安装文件夹中运行python faceswap.py extract。这将从src文件夹拍摄照片并将面部提取到extract文件夹中。
从您的安装文件夹中运行python faceswap.py train。这将从包含两张脸的照片的两个文件夹中拍摄照片，并训练将保存在models文件夹内的模型。
从您的安装文件夹中运行python faceswap.py convert。这将从original文件夹中拍摄照片并将新面孔应用到modified文件夹中。
您可以通过运行来运行GUI python faceswap.py gui
换脸/换头
https://github.com/iperov/DeepFaceLab
https://radek350.wordpress.com/2018/02/17/myfakeapp-fakeapp-alternative/
https://github.com/sunattic/AISuperstar
https://github.com/joshua-wu/deepfakes_faceswap
https://github.com/llSourcell/deepfakes
https://github.com/dfaker/df
https://github.com/gsurma/face_generator

CTR
https://github.com/shenweichen/DeepCTR

经过预先训练的30多种语言的单词向量
https://github.com/hcxiong/wordvectors
人名、地址、邮箱、手机号、手机归属地 等信息的抽取，rake短语抽取算法。
pip3 install cocoNLP
清华大学XLORE:中英文跨语言百科知识图谱
https://xlore.org/ttl/xlore.all.zip

智能家居
https://github.com/apanly/piRobot
https://github.com/apanly/autohome
https://github.com/2shou/TextGrocery.git #短文本分类工具
BeautifulSoup(HTML/XML的解析器)
http://www.pm25.in/api_doc
https://www.faceplusplus.com.cn/
pip3 install jieba tgrocery

远程控制玩具车
https://github.com/pjq/rpi

sudo apt-get install libttspico-utils
https://github.com/GwadaLUG/pico-read-speaker

• libttspico-data (https://openrepos.net/content/mickaelh/libttspico-data)
• libttspico0 (https://openrepos.net/content/mickaelh/libttspico0)
• libttspico-utils (https://openrepos.net/content/mickaelh/libttspico-utils)
• libttspico-dev (https://openrepos.net/content/mickaelh/libttspico-dev)
  or
• sudo apt-get install libttspico0 libttspico-utils libttspico-data
  wget https://raw.githubusercontent.com/stevenmirabito/asterisk-picotts/master/picotts-install.sh -O - | sh
• svox (pico2wave) https://packages.debian.org/source/squeeze/svox
  https://github.com/mscdex/speaky
  https://github.com/grigi/talkey
  https://pbxinaflash.com/community/threads/svox-pico-tts-for-asterisk.17859/
  pico2wave -l fr-FR -w /tmp/test.wav “Ceci est un test”
  aplay /tmp/test.wav

https://github.com/zaf/asterisk-googletts
brew install sox mpg123 pulseaudio espeak
soxi sox play
play existing-file.wav
sox existing-file.wav −d

https://www.google.com.hk/speech-api/v1/recognize?xjerr=1&client=chromium&pfilter=2&lang=zh-CN&maxresults=6
https://github.com/apanly/piRobot/blob/master/stt/google.py

cd /
wget http://incrediblepbx.com/picotts.tar.gz
tar zxvf picotts.tar.gz
cd /root
./picotts-install.sh
sed -i ‘s|en)|en-US)|’ /etc/asterisk/extensions_custom.conf
sed -i ‘s|googletts|picotts|’ /etc/asterisk/extensions_custom.conf
asterisk -rx “dialplan reload”

espeak --stdout “this is a test” | paplay
echo “these are my notes” > text.txt
espeak --stdout -f text.txt > text.wav
paplay text.wav # you should hear “these are my notes”
play text.wav

基于STM32的孤立词语音识别
https://github.com/gk969/stm32-speech-recognition
http://gk969.com/stm32-speech-recognition/

图灵
聊天
http://www.tuling123.com/openapi/api

图片识别文字
pip3 install baidu-aip
https://github.com/shuoGG1239/Image2Text
https://github.com/lancezhange/smoke_recognition 图片烟雾识别
python3 smokeDetection.py
pip3 install pytesseract

https://www.cnblogs.com/wzben/p/5930538.html
brew install --with-training-tools --all-languages tesseract
https://github.com/tesseract-ocr/tessdata
https://github.com/tesseract-ocr/tessdata/tree/3.04.00
https://github.com/tesseract-ocr/tessdata_fast/

tesseract -v tesseract --list-langs 查看版本+语音
tesseract 图片名称 生成的结果文件的名称 字库
tesseract test.jpg result -l chi_sim
tesseract -l chi_sim+eng
tesseract 1234.png 1234 -l chi_sim -psm 6
tesseract --help-psm
0 定向脚本监测（OSD）
1 使用OSD自动分页
2 自动分页，但是不使用OSD或OCR（Optical Character Recognition，光学字符识别）
3 全自动分页，但是没有使用OSD（默认）
4 假设可变大小的一个文本列。
5 假设垂直对齐文本的单个统一块。
6 假设一个统一的文本块。
7 将图像视为单个文本行。
8 将图像视为单个词。
9 将图像视为圆中的单个词。

将图片转换成tif格式，用于后面生成box文件。可以通过画图，然后另存为tif即可
[lang].[fontname].exp[num].tif
生成box文件
tesseract mjorcen.normal.exp0.jpg mjorcen.normal.exp0 -l chi_sim batch.nochop makebox
box文件和对应的tif一定要在相同的目录下，不然后面打不开。
打开jTessBoxEditor矫正错误并训练 打开train.bat
tesseract mjorcen.normal.exp0.jpg mjorcen.normal.exp0 nobatch box.train
unicharset_extractor mjorcen.normal.exp0.box
新建一个font_properties文件
里面内容写入 normal 0 0 0 0 0 表示默认普通字体
shapeclustering -F font_properties -U unicharset mjorcen.normal.exp0.tr
mftraining -F font_properties -U unicharset -O unicharset mjorcen.normal.exp0.tr
cntraining mjorcen.normal.exp0.tr
最后会生成五个文件，把目录下的unicharset、inttemp、pffmtable、shapetable、normproto这五个文件前面都加上normal.
combine_tessdata normal.
得到训练好的字库。
把 normal.traineddata 复制到Tesseract-OCR 安装目录下的tessdata文件夹中
tesseract mjorcen.normal.exp0.jpg mjorcen.normal.exp0 -l normal

素材合成，(多个素材合成)
打开jTessBoxEditor工具，菜单栏：tools->Merge TIFF…，选中要合成的图片并保存为为：huiyi.fitt。
生成box文件
tesseract huiyi.tif huiyi -l chi_sim -psm 10 batch.nochop makebox
执行后会在生成一个名为huiyi.box的box文件。
用文本编辑器或者xcode打开编辑保存
生成.tr文件
tesseract huiyi.tif huiyi -psm 10 nobatch box.train
生成unicharset文件
unicharset_extractor huiyi.box

jTessBoxEditor
https://sourceforge.net/projects/vietocr/files/jTessBoxEditor/
java -Xms4096m -Xmx4096m -jar jTessBoxEditor.jar

图片转tiff
pip3 install tifffile
python3 /usr/local/lib/python3.7/site-packages/tifffile/tifffile.py --help
vi ~/.bash_profile
alias tifffile=‘python3 /usr/local/lib/python3.7/site-packages/tifffile/tifffile.py’
source ~/.bash_profile
tifffile --help

go get -u github.com/brunsgaard/img2tiff
cd $GOPATH/src/github.com/brunsgaard/img2tiff

https://blog.csdn.net/qq_25806863/article/details/67637567
vi process-tessdata.sh
#!/bin/sh
read -p “输入你语言:” lang
echo ${lang}
read -p “输入你的字体:” font
echo ${font}
echo “所以完整文件名为：”
echo $lang.${font}.exp0.tif
echo “开始。。。”
echo ${font} 0 0 0 0 0 >font_properties
tesseract $lang.${font}.exp0.tif $lang.${font}.exp0 nobatch box.train
unicharset_extractor $lang.${font}.exp0.box
shapeclustering -F font_properties -U unicharset $lang.${font}.exp0.tr
mftraining -F font_properties -U unicharset -O unicharset $lang.${font}.exp0.tr
cntraining $lang.${font}.exp0.tr
echo “开始重命名文件”
mv inttemp ${font}.inttemp
mv normproto ${font}.normproto
mv pffmtable ${font}.pffmtable
mv shapetable ${font}.shapetable
mv unicharset ${font}.unicharset
echo “生成最终文件”
combine_tessdata ${font}.
echo “完成”

识别车牌
https://github.com/zeusees/HyperLPR
pip install hyperlpr

CNN的OCR车牌识别
https://github.com/huxiaoman7/mxnet-cnn-plate-recognition

一款入门级的人脸、视频、文字检测以及识别的项目.
https://github.com/vipstone/faceai
https://github.com/bairdzhang/smallhardface

pip3 install dlib
训练模型用于是人脸识别的关键，用于查找图片的关键点。
wget http://dlib.net/files/shape_predictor_68_face_landmarks.dat.bz2
当然你也可以训练自己的人脸关键点模型，这个功能会放在后面讲。
下载好的模型文件，我的存放地址是：C:\Python36\Lib\site-packages\dlib-data\shape_predictor_68_face_landmarks.dat.bz2
解压：shape_predictor_68_face_landmarks.dat.bz2得到文件：shape_predictor_68_face_landmarks.dat
https://github.com/hcxiong/faceai/blob/master/doc/detectionDlib.md
https://github.com/hcxiong/faceai/blob/master/doc/videoOpenCV.md
https://github.com/hcxiong/faceai/blob/master/doc/videoDlib.md
https://github.com/hcxiong/faceai/blob/master/doc/faceRecognitionOutline.md

人脸检测
https://github.com/610265158/DSFD-tensorflow

https://github.com/kpzhang93/MTCNN_face_detection_alignment
https://github.com/ydwen/caffe-face
https://github.com/deepinsight/insightface
https://github.com/deepinsight/insightface/wiki/Model-Zoo
MS1M-Arcface
https://pan.baidu.com/s/1S6LJZGdqcZRle1vlcMzHOQ
https://github.com/Linzaer/Ultra-Light-Fast-Generic-Face-Detector-1MB

https://github.com/ZhaoJ9014/face.evoLVe.PyTorch
https://github.com/Cadene/pretrained-models.pytorch

2012年视觉对象课程挑战（VOC2012）
http://host.robots.ox.ac.uk:8080/pascal/VOC/voc2012/index.html

https://github.com/fighting41love/funNLP
中英文敏感词过滤 https://github.com/observerss/textfilter
97种语言检测 https://github.com/saffsd/langid.py
另一个语言检测https://code.google.com/archive/p/language-detection/
中国手机归属地查询 https://github.com/ls0f/phone
phone国际手机、电话归属地查询 https://github.com/AfterShip/phone
根据名字判断性别： https://github.com/observerss/ngender
人名语料库 https://github.com/wainshine/Chinese-Names-Corpus
中文缩写库 https://github.com/zhangyics/Chinese-abbreviation-dataset/blob/master/dev_set.txt
汉语拆字词典 https://github.com/kfcd/chaizi
词汇情感值 https://github.com/rainarch/SentiBridge/blob/master/Entity_Emotion_Express/CCF_data/pair_mine_result
中文词库、停用词、敏感词 https://github.com/dongxiexidian/Chinese
汉字转拼音 https://github.com/mozillazg/python-pinyin
中文繁简体互转 https://github.com/skydark/nstools/tree/master/zhtools
英文模拟中文发音引擎 funny chinese text to speech enginee https://github.com/tinyfool/ChineseWithEnglish
同义词库、反义词库、否定词库 https://github.com/phunterlau/wangfeng-rnn
无空格英文串分割、抽取单词 https://github.com/keredson/wordninja
结巴中文分词 https://github.com/fxsjy/jieba
百度中文词法分析（分词+词性+专名）系统 https://github.com/baidu/lac
https://github.com/baidu/AnyQ 百度FAQ自动问答系统
https://github.com/baidu/Senta 百度情感识别系统
Scattertext 文本可视化:https://github.com/JasonKessler/scattertext
中文字符数据:https://github.com/skishore/makemeahanzi
中文识别
https://github.com/breezedeus/cnocr
python3 scripts/cnocr_predict.py --file multi-line_cn1.png
https://github.com/diaomin/crnn-mxnet-chinese-text-recognition

语料
https://github.com/codemayq/chaotbot_corpus_Chinese
https://github.com/gunthercox/chatterbot-corpus
https://github.com/MarkWuNLP/MultiTurnResponseSelection
https://github.com/wb14123/couplet-dataset
中文古诗自动作诗机器人
https://github.com/jinfagang/tensorflow_poems
python3 train.py
python3 compose_poem.py
python3 main.py -w poem --no-train

基於向量匹配的情境式聊天機器人
https://github.com/zake7749/Chatbot
https://github.com/zake7749/PTT-Chat-Generator
用于主题建模，文档索引 和大型语料库的相似性检索。目标受众是 自然语言处理（NLP）和信息检索（IR）社区。
https://github.com/RaRe-Technologies/gensim
PTT 八卦版問答中文語料
https://github.com/zake7749/Gossiping-Chinese-Corpus

处理中文文本内容
https://github.com/isnowfy/snownlp
文本相似度
https://github.com/seatgeek/fuzzywuzzy
https://github.com/sloria/TextBlob

ocr
http://apis.baidu.com/apistore/idlocr/ocr
https://github.com/deloz/baiduocr
https://github.com/tesseract-ocr/tesseract
brew install --with-training-tools --all-languages tesseract
brew install imagemagick
tesseract imagename outputbase [-l lang] [–oem ocrenginemode] [–psm pagesegmode] [configfiles…]
tesseract -l chi_sim data/test_data.png out_test_data
chi_sim.traineddata
eng.traineddata
https://github.com/naptha/tesseract.js
http://tesseract.projectnaptha.com/
https://github.com/madmaze/pytesseract
https://github.com/thiagoalessio/tesseract-ocr-for-php
https://github.com/otiai10/gosseract
https://github.com/Greedysky/TTKOCR
https://github.com/Aixtuz/CardScanner
https://github.com/iChenwin/pytesseractID
https://github.com/csxiaoyaojianxian/BloodTestReportOCR
https://github.com/bigchao8/Opencv-ImageBase
基于caffe
https://github.com/JinpengLI/deep_ocr
python reco_chars.py
实现ctpn+crnn+ctc实现不定长场景文字OCR识别
https://github.com/xiaofengShi/CHINESE-OCR
环境部署
Bash
##GPU环境
sh setup.sh
##CPU环境
sh setup-cpu.sh
##CPU python3环境
sh setup-python3.sh
使用环境：python3.6+tensorflow1.7+cpu/gpu
https://github.com/jimmyleaf/ocr_tensorflow_cnn
安装
http://caffe.berkeleyvision.org/install_osx.html
brew tap homebrew/science
brew install hdf5 opencv

文字识别
http://www.robots.ox.ac.uk/~vgg/data/text/
https://yq.aliyun.com/articles/109555?t=t1
https://github.com/YCG09/chinese_ocr
sh setup.sh #环境部署
python demo.py #Demo 将测试图片放入test_images目录，检测结果会保存到test_result中
训练 数据集：https://pan.baidu.com/s/1QkI7kjah8SPHwOQ40rS1Pw (密码：lu7m) 图片解压后放置到train/images目录下，描述文件放到train目录下
cd train
python train.py
https://github.com/JarveeLee/SynthText_Chinese_version
https://github.com/Belval/TextRecognitionDataGenerator
https://github.com/Sanster/text_renderer

用keras实现OCR定位、识别
https://github.com/xiaomaxiao/keras_ocr

https://github.com/eragonruan/text-detection-ctpn 文字区域检测CTPN
https://github.com/eragonruan/text-detection-ctpn/releases
python ./ctpn/demo_pb.py
cd lib/utils
chmod +x make.sh
./make.sh
prepare data
cd lib/prepare_training_data
python split_label.py
it will generate the prepared data in current folder, and then run
python ToVoc.py
python ./ctpn/train_net.py

主流ocr算法研究实验性的项目，目前实现了CNN+BLSTM+CTC架构
https://github.com/senlinuc/caffe_ocr
https://github.com/isee15/Card-Ocr
基于caffe的LSTM OCR案例，能够利用该案例完成序列的识别，包括验证码、车牌、身份证号码、地址等长序列动长的内容识别
https://github.com/dlunion/CaffeLSTM-OCR

物体识别
https://github.com/open-mmlab/mmdetection
https://github.com/HRNet/HRNet-Object-Detection

CTC可以被用来训练端对端的语音识别系统
https://github.com/baidu-research/warp-ctc
git clone https://github.com/baidu-research/warp-ctc.git
cd warp-ctc
mkdir build
cd build
cmake …/
make
http://ilovin.me/2017-04-06/tensorflow-lstm-ctc-ocr/
https://github.com/ilovin/lstm_ctc_ocr

https://github.com/bgshih/crnn
https://github.com/meijieru/crnn.pytorch

腾讯优图OCR云平台识别身份证、银行卡、行驶证、驾驶证，依赖包小，识别次数免费
https://api.youtu.qq.com/youtu/ocrapi/
https://open.youtu.qq.com/#/open
https://github.com/Tencent-YouTu/nodejs_sdk
https://github.com/Tencent-YouTu/Python_sdk
https://github.com/Tencent-YouTu/Go_sdk
https://github.com/TencentYouTu/ios_sdk
https://github.com/TencentYouTu/android_sdk

基于Xception的腾讯验证码识别（样本+代码）
https://github.com/bojone/n2n-ocr-for-qqcaptcha
10万验证码样本公开如下：
链接: https://pan.baidu.com/s/1mhO1sG4 密码: j2rj
https://github.com/keras-team/keras

百度莱茨狗抢购脚本
https://github.com/Acamy/pet-chain-buyer
https://github.com/yanwii/pet-chain
https://pet-chain.baidu.com/

图片验证码识别
https://www.showapi.com/api/view/184
https://github.com/Yaoshicn/decaptcha
https://github.com/dingyaguang117/ImageRecognizeOf58
https://github.com/CrazyHusen/IdentificationCodes

百度二代身份证识别
https://github.com/DophinL/baidu-ocr-idcard
https://github.com/Freeza91/baidu_ocr
百度OCR文字识别API For Ruby Gems
https://rubygems.org/gems/baidu_ocr
https://aip.baidubce.com/rest/2.0/ocr/v1/general
https://aip.baidubce.com/rest/2.0/ocr/v1/general_basic
https://aip.baidubce.com/rest/2.0/ocr/v1/general_enhanced
https://aip.baidubce.com/rest/2.0/ocr/v1/accurate_basic
https://aip.baidubce.com/rest/2.0/ocr/v1/accurate
https://aip.baidubce.com/rest/2.0/ocr/v1/bankcard
https://aip.baidubce.com/rest/2.0/ocr/v1/idcard
https://aip.baidubce.com/rest/2.0/ocr/v1/webimage
https://aip.baidubce.com/rest/2.0/ocr/v1/driving_license
https://aip.baidubce.com/rest/2.