AI-Speaker

Implemented Artificial Intelligent Speaker.

Used API

Speech to Text : Google Streaming Speech Recognition API, Google Speech Recognition
Conversaion : Watson, Aibril(korean)
Text to Speech : gTTS, Naver Clova TTS, AWS Polly TTS

Requirements

• Ubuntu 16.04
• Python 3.5

Dependencies

• sudo apt-get install python3-dev virtualenv
• sudo apt-get install portaudio19-dev
• sudo apt-get install ffmpeg

Setting Environment

Google Streaming Speech Recognition

export GOOGLE_APPLICATION_CREDENTIALS=/path/to/your/credentials-key.json

Aibril Conversation

export watson_username=YOUR_USER_NAME
export watson_password=YOUR_PASSWORD
export watson_workspace=YOUR_WORKSPACE
export watson_url=YOUR_URL
export watson_version=YOUR_VERSION

Naver Clova TTS

export naver_tts_client_id=YOUR_CLIENT_ID
export naver_tts_client_secret=YOUR_CLIENT_SECRET

AWS Polly TTS

• https://github.com/yonghankim/aws-polly#3-getting-started-with-the-aws-cli

Getting Started

Creating virtualenv

1. cd AI-Speaker
2. virtualenv env --python=python3.5
3. source env/bin/activate
4. pip install -r requirements.txt

Run

• python main.py