say-again

S3-memoized Text-to-Speech adapter for sayings worth repeating. Default TTS is AWS Polly.

Installation

npm install --save say-again

AWS credentials (JSON file)

Configure a SayAgain AWS configuration file (e.g., MY_AWS.json)

cp test/data/aws-sample.json MY_AWS.json

Edit MY_AWS.json with your actual AWS credentials.

AWS credentials (environment)

You can also configure AWS credentials using nodejs environment variables:

• aws_config_region
• aws_config_accessKeyId
• aws_config_secretAccessKey

Example

const CFGPATH = "MY_AWS.json";
const { SayAgain, TtsPolly } = require("say-again");

var request = {
  "api": "aws-polly",
  "apiVersion": "v4",
  "audioFormat": "mp3",
  "voice": "Amy",
  "language": "en-GB",
  "text": "<prosody rate=\"-30%\" pitch=\"-10%\">hello</prosody>"
}

var v = new SayAgain(CFGPATH);
var res = await v.speak(request);
// { 
//   request, 
//   s3Key, 
//   response: {
//     mime: "audio/mpeg",
//     base64: "..."
//   }
// }

To generate an MP3 file from the result:

var buf = Buffer.from(res.response.base64, 'base64');
fs.writeFileSync('hello.mp3', buf);