Amazon Polly TTS Engine for Windows

Use Amazon Polly voices in native Windows TTS applications

Installation QuickStart

Step 1: Install the AWS Client (if it's not already installed on your machine)

Follow the instructions on this page to install the AWS Command Line Interface on Microsoft Windows.

IMPORTANT: If you're not sure what to do here, follow the instructions in the MSI Installer section.

Step 2: Setup your AWS user account

1. Login to your AWS account

2. After you've logged in, click Services from the top menu bar, then type iam in the search box. Click IAM when it pops up.

3. On the left, click Users

4. Click Add User

5. Type in polly-windows-user (you can use any name)

6. Click the Programmatic access checkbox and leave AWS Management Console access unchecked

7. Click Next: Permissions

8. Click Attach existing policies directly

9. At the bottom of the page, in the search box next to Filter: Policy type, type polly

10. Click the checkbox next to AmazonPollyFullAccess

11. Click Next: Review

12. Click Create user

    IMPORTANT: Don't close the web page. You'll need both the Access key ID and the Secret access key in the next step.

Step 3: Configure the AWS Client

Amazon Polly TTS Engine for Windows requires an AWS profile called polly-windows. It's in the roadmap to allow custom profiles.

1. Open a Windows command prompt.

2. Type this command:

     aws configure --profile polly-windows
   

3. For AWS Access Key ID and AWS Secret Access Key, use the values from the previous step.

4. For Default region name, you can hit Enter for the default (us-east-1) or enter a different region. Make sure to use all lower-case. This is case-sensitive.

5. For Default output format, just hit Enter

6. Verify this worked by running this command. You should see a list of voices:

     aws --profile polly-windows polly describe-voices
   

Step 4: Install Amazon Polly TTS Engine for Windows

Download and run the installer: 32-bit or 64-bit

Verify that the installer worked by opening Control Panel and go to Change text to speech settings. In the Voice selection drop-down, you should see all of the Amazon Polly voices. Picking a voice will automatically play a sample.

Adobe Captivate Support

Even though there is a drop-down voice selection in the Audio / Speech Management window, apparently Adobe Captivate only uses the default voice you choose in Windows Control Panel. The voice selection in Captivate is completely ignored.

To get around this, you can manually specify the voice name that you want to use by using the voice tag.

Instructions for Manually Specifying a Voice in Adobe Captivate

1. Set the default Windows voice in Control Panel to any of the Amazon Polly voices.
2. Start Adobe Captivate and open any of your previous documents.
3. Choose Speech Management from the Audio menu
4. (See the screenshot below): Where you enter the text to speak, put the highlighted yellow tags around the text, replacing the name with the voice name you want to use. For example, <speak><voice name="Matthew">Hi, this is Matthew</voice></speak> will automatically speak the phrase Hi, this is Matthew in the Matthew voice and <speak><voice name="Ivy">Hi, this is Ivy</voice></speak> will say Hi, this is Ivy in the Ivy voice.

Click here for the complete list of Polly voice names.

IMPORTANT: It doesn’t matter which voice you select at the top or the left side menu, as long as it’s a Polly voice. The tag will override those selections. However, if you choose a non-Polly voice as the Windows default voice, then the tag will NOT work with the Polly voices.

Also, you can only use one <speak> tag per block of text, and you can only use one <voice> tag per <speak> tag (e.g., you can’t do <speak><voice name="ivy">I’m Ivy.</voice><voice name="Matthew">I’m Matthew</voice></speak>).

Setting Up Your Development Environment

Overview

There are four components to the SAPI plugin:

TTSEngine

This is the main source code for the actual driver itself.

InstallVoices

Command-line tool that registers the voices.

Polly Player

Simple tool to test out the voices

Installer

Installation tool that installs TTSEngine, InstallVoices and Polly Player

Setting up TTSEngine/InstallVoices

1. Install Visual Studio 2019 with C++ desktop support

2. Clone this repo:

   > git clone https://github.com/awslabs/amazon-polly-tts-for-windows
   

3. Install [vcpkg](vcpkg: a C++ package manager for Windows, Linux, and macOS | Microsoft Docs):

   NOTE: This step may take up to an hour to complete.

   > cd amazon-polly-tts-for-windows
   > .\install-vcpkg.bat
   

4. Start Visual Studio.

5. Click File/Open Folder and select the amazon-polly-tts-for-windows folder.

Setting up Polly Player

1. Install Visual Studio 2019 with C# support

2. Start Visual Studio.

3. Open PollyPlayer\PollyPlayer.csproj

Setting up the Installer

• NOTE: x86 and x64 builds have separate installer files.

1. Install innosetup

2. Start Innosetup.

3. Open PollyTTSInstaller-x64.iss or PollyTTSInstaller-x86.iss