/text-to-speech-polling-api-java-example

Primary LanguageJavaMIT LicenseMIT

Narakeet Text To Speech Polling (Long content) API example in Java

This repository provides a quick example demonstrating how to access the Narakeet Narakeet Long Content API from Java.

The long content API is suitable for large audio conversion tasks, and can produce professional quality uncompressed WAV files using realistic text to speech.

Note that Narakeet also has a simpler text to speech API, suitable for smaller conversion tasks, that directly streams back the results. See the Streaming Text to Speech API Example for more information on how to use that.

The example sends a request to generate an audio file, then downloads the resulting audio into a local file.

Prerequisites

To use this example, you will need Java 11 or later, and Maven 3 or later, and an API key for Narakeet.

The example uses the org.apache.http library to execute HTTPS requests.

Running the example

  1. Set a local environment variable called NARAKEET_API_KEY, containing your API key (or modify src/main/java/com/narakeet/Example.java line 8 to include your API key).
  2. optionally modify src/main/java/com/narakeet/Example.java lines 11 to change the voice, format and text sent to the text to speech engine.
  3. run mvn install to download the dependencies
  4. run mvn compile exec:java to run the conversion.

Running the example from Docker

For easy development, this repository also includes a Docker container with Amazon Correto 8, which can be used to run the conversion without installing Java locally. To use Docker:

  1. run make init to create the Docker image locally
  2. run make bootstrap to download Maven dependencies
  3. run make execute NARAKEET_API_KEY=<YOUR API KEY> to produce the result file.
  4. check the output subdirectory to play the downloaded file

More information

Check out https://www.narakeet.com/docs/automating/text-to-speech-api/ for more information on the Narakeet text to speech API.