raspberry pi - Python 3 Artificial Intelligence: Offline STT and TTS -
so have been programming python awhile now. have made simple ai chatbots in python communicate via text. want move next level, kind of personal companion ai. goal put on raspberry(i have portable charger, microphone, , speaker compatible pi) , make offline ai talking to, taking notes, remembering info, etc. know way incorporate offline stt , tts engines python program. (most of stt , tts engines i've found online via google, amazon, etc.) in advance.
i have check offline stt. tried run these below , see comments.
you can @ them according purposes.
online
wit.ai https://wit.ai/ can used in commercial product.
build brand-unique, natural language interactions bots, applications, services, , devices. https://api.ai/ https://docs.api.ai/docs/languages
offline
cmusphinx http://cmusphinx.sourceforge.net cmu sphinx speech recognition engines. cmu sphinx - speech recognition toolkit - offline speech recognition, due low resource requirements can used on mobile.
https://pypi.python.org/pypi/speechrecognition/ https://github.com/uberi/speech_recognition
espeak ng open source speech synthesizer supports 93 languages , accents. (supports chineese) https://github.com/rhdunn/espeak espeak compact, multi-language, open source text-to-speech synthesizer. reads text files , stdin ( in console-line).
uwp speech recognition microsoft universal windows platform https://docs.microsoft.com/en-us/windows/uwp/input-and-devices/speech-recognition enables speech recognition command , control within windows runtime app.
kaldi toolkit speech recognition written in c++ , licensed under apache license v2.0. kaldi intended use speech recognition researchers. (supports chineese) https://github.com/kaldi-asr/kaldi http://kaldi-asr.org/doc/about.html during installation on windows, got errors , couldn’t continue. in addition, written on web site: “note windows setup becoming out of date , not regularly tested, , not code compiles on it.”
tensorflow-speech-recognition https://github.com/pannous/tensorflow-speech-recognition speech recognition using google's tensorflow deep learning framework, sequence-to-sequence neural networks. research oriented. project carried out https://github.com/pannous/caffe-speech-recognition
deep speech https://github.com/mozilla/deepspeech tensorflow implementation of baidu's deepspeech architecture research oriented.
Comments
Post a Comment