Text-To-Speech telegram bot implementation. Randomly choices voice of speaker taking into account the language of a given text.
User sends texts to the telegram-bot. Dispatcher receives and pushes update to the rabbitmq queue. TTS-service listens to the queue, pulls update out there and sends text from this update to aws-polly to process. After audio is received, TTS-service sends it to the s3-storage and pushes url to rabbitmq queue which is listened by dispatcher. Dispatcher responds to user, sending him this file.