This is a project that java, kafka and docker technologies are used together.
Here I try to demonstrate how to use java, kafka and docker to process a stream of voice data.
We use AssemblyAI library to convert auido data from byte array to text (You should change the API key inside SpechToTextClient.java. Existing one won't work)
NOTE: Translate service are just mock classes. and Google Cloud Translation API can be used instead of mock class.
├───kafka-streams-voice-command-consumer
├───kafka-streams-voice-command-parser
├───kafka-streams-voice-command-producer
└───kafka-voice-command-common
kafka-streams-voice-command-consumer: This system connects to kafka topic and print the message on the console.
kafka-streams-voice-command-producer: This system connects to kafka topic and produce voice events.
kafka-streams-voice-command-parser: This system connects to kafka, consume events from voice-commands topic, process and save them to recognized-commands or unrecognized-commands topics.
kafka-voice-command-common: This is a library app that contains common models, serdes objects and configuration utils.
docker-compose file is available for you to try it and see the result on you machine.
