transientDocument Id generating API does not work when filestream is passed from S3

If I create a readFileStream from local directory then I am able to upload template and get transient Id but If done the same by fetching the uploaded docx file in S3 it doesn't work. I am getting the file Stream using the below code and it doesn't work.

async function getFileStream() {
  // configuring the AWS environment
  AWS.config.update({
    accessKeyId: process.env.AWS_ACCESS_KEY,
    secretAccessKey: process.env.AWS_SECRET_KEY,
  });
  // initialize s3
  const s3 = new AWS.S3();
  const file = s3.getObject({ Bucket: 'bucket', Key: 'key.docx' }).createReadStream();
  const pass = new stream.PassThrough();

  stream.pipeline(
    file,
    pass,
    (err) => {
      if (err) {
        console.error(err);
      }
    },
  );
  return file;
}

 
But when I implemented it directly like below then it works perfectly.

const file = fs.createReadStream(file.path);
return file;

There is no error in the above case but the response never comes until 120 seconds and then it throws timeout error.