This project is a proxy to provide support for real-time streaming functionality for Open Speech API https://open-speech-ekstep.github.io/ from browsers or any clients that doesn't support gprc bi-directional streaming. Refer the real-time streaming documentation for architecture details https://open-speech-ekstep.github.io/asr_streaming_service/.

• Getting Started
• Getting Help

npm i 

• Add model hosted address in language_map.json file in the project root. Example:

  {
      "<ip-address/domain>:<port>": [
          "hi",
          "en"
      ],
      "<ip-address/domain>:<port>": [
          "ta",
          "te"
      ]
  }

• Set the folder path of language_map.json as env variable: config_base_path="<project_root_folder_path>" (eg: /users/node/speech-recognition-open-api-proxy)

npm start

This service url can be used in speech-recognition-open-api-client for real-time inferencing experience.

1. Port of the service can be changed using PORT environment variable.

The best way to interact with our team is through GitHub. Please open an issue in github.