For blind and visually impaired computer users, Apple’s Alex is widely considered one of the most iconic, responsive, and natural-sounding Text-to-Speech (TTS) voices ever created. If you have been looking for an easy way to use the Apple Alex voice on Windows, this complete guide is for you. In this tutorial, you will learn step-by-step how to set up Alex voice on NVDA using Panthera speech, configure Alex voice on JAWS with SAPI 5, and even upgrade to the crystal-clear Sequoia Alex voice.
Supported Mac OS X Speech Engines & Voices
Through the Panthera Speech project, you are not restricted to just one macOS version. You can choose and install any of the following speech engines based on your personal preference:
Tiger Speech (Mac OS X 10.4): Contains classic MacinTalk novelty voices (Fred, Victoria, Bruce, Agnes, Albert, Ralph, Zarvox, etc.).
Leopard Speech (Mac OS X 10.5): The very first version to introduce the revolutionary Alex voice, alongside the classic legacy voices.
Snow Leopard Speech (Mac OS X 10.6): Optimized MacinTalk engine with the Alex voice and improved performance.
Lion Speech (Mac OS X 10.7): The most stable and updated legacy engine for Alex and classic voices. It also serves as the host engine if you want to load newer voice data from macOS Sequoia.
Note: You can install one specific engine or install all of them side by side.
How to Use Alex & Mac Voices with NVDA (Step-by-Step)
Setting up Mac voices on NVDA is straightforward using the Panthera Speech add-on.
Step 1: Install the Panthera Speech Add-on
- Download the pantheraspeech.nvda-addon file from the Download Links section below.
- Press Enter on the downloaded .nvda-addon file.
- When NVDA asks for installation confirmation, select Yes (or press Alt + Y).
- Restart NVDA when prompted.
Step 2: Download Your Preferred Voice Pack
Because the add-on acts as an engine loader, the speech data files must be downloaded separately:
Choose the version you want to install from the download section:
tiger.zip
leopard.zip
snowleopard.zip
lion.zip (Recommended if you want Alex)
Extract the downloaded zip file using 7-Zip, WinRAR, or Windows File Explorer.
Open the extracted folder (e.g., if you extracted lion.zip, open the lion folder; if leopard.zip, open leopard).
Press Control + A to select all files and folders inside, then press Control + C to copy them.
join our telegram channel for more interesting updates
Step 3: Paste Speech Data into the Corresponding NVDA Folder
- Open the NVDA Menu by pressing NVDA + N.
- Navigate to Tools > Mac OS X speech data.
- Select the engine corresponding to the voice pack you downloaded:
- Tiger speech — Mac OS X 10.4
- Leopard speech — Mac OS X 10.5
- Snow Leopard speech — Mac OS X 10.6
- Lion speech — Mac OS X 10.7
- Press Tab to reach Open the speech data folder, then press Enter.
- Once the folder opens in File Explorer, press Control + V to paste all copied files.
- (Important: Paste the contents of your extracted folder directly here).
- Restart NVDA by pressing NVDA + Q, selecting Restart, and pressing Enter.
- (Alternative Manual Path: You can also find your voice folders in %appdata%\nvda\macintalk).
Step 4: Select Your Synthesizer in NVDA
- Press NVDA + Control + S to open the Synthesizer selection dialog.
- In the list, choose the engine you installed (e.g., Lion Speech (Alex, MacinTalk 4.0), Leopard Speech, etc.).
- Press Enter.
- Press NVDA + Control + V to customize your speech rate, pitch, volume, and choose individual voices (Alex, Bruce, Victoria, etc.).
Surprise!
Now that you have the Alex voice running smoothly on your screen reader, why stop there? If you love high-quality text-to-speech tools, AI technology, and uninterrupted entertainment, we have great news for you.
Get genuine access to ElevenLabs Creator, ChatGPT Plus / Claude AI, Spotify Premium, YouTube Premium, Netflix, and many more popular subscriptions at highly nominal and affordable prices—without breaking your bank!
Checkout to Explore Our Discounted Subscriptions & Deals
How to Upgrade to High-Quality Sequoia Alex Voice
The Alex voice included with Mac OS X 10.7 Lion is fully functional, but if you want higher-resolution acoustic sampling, you can use the voice data from macOS Sequoia inside the Lion Speech engine.
Step 1: Download & Rename the Sequoia Voice Folder
- Download the sequoia-alex voice package from the download section.
- Extract the zip file and find the folder named sequoia-speech.
- Rename this folder to:
SpeechVoice - Copy the renamed Alex.SpeechVoice folder (Control + C).
Step 2: Switch to a Different Synthesizer Temporarily
- Press NVDA + Control + S.
- Temporarily switch to another synthesizer (e.g., Windows OneCore or eSpeak NG) and press Enter.
- (This unlocks the speech files so Windows allows you to replace them).
Step 3: Replace the Lion Voice Data
- Press NVDA + N > Tools > Mac OS X speech data.
- Select Lion speech — Mac OS X 10.7 and press Enter on Open the speech data folder.
- Inside the folder, open:
Speech -> Voices - Find the original Alex.SpeechVoice folder and rename it as a backup:
SpeechVoice-Lion-Backup - Press Control + V to paste your new Sequoia Alex.SpeechVoice folder into this directory.
Step 4: Reactivate Lion Speech
- Restart NVDA (NVDA + Q -> Restart).
- Press NVDA + Control + S, select Lion Speech (Alex, MacinTalk 4.0), and press Enter.
Your screen reader will now speak using the modern macOS Sequoia Alex voice model powered by the Lion engine.
How to Use Alex Voice with JAWS Screen Reader (via Panthera SAPI)
JAWS requires speech synthesizers to run natively or via Microsoft SAPI 5. The Panthera SAPI driver allows you to use Alex and all classic Mac voices inside JAWS.
Step 1: Install Panthera SAPI
- Download panthera-sapi-setup.exe from the download section.
- Run the installer and complete the setup wizard.
- Step 2: Extract Voice Data to AppData
- Download your preferred voice package (lion.zip, leopard.zip, tiger.zip, etc.).
- Extract the zip file and copy the extracted folder.
- Press Windows Key + R to open the Run dialog.
- Type %appdata% and press Enter.
- Locate or create the folder named:
macintalk-data - Paste your copied voice files inside %appdata%\macintalk-data.
- (If you want the high-quality Sequoia Alex voice, replace the Alex.SpeechVoice inside macintalk-data\Speech\Voices using the same method described in Part 2).
Step 3: Register the Voices in Panthera SAPI Settings
- Launch Panthera SAPI Settings:
- Search for Panthera SAPI in the Windows Start Menu, OR
- Go to C:\Program Files\Panthera SAPI and run panthera_settings.exe.
- When the utility detects your voice files, a prompt will appear:
- “Do you want to register the voice?”
- Click Yes to register Alex and all included Mac voices into Windows SAPI 5.
Step 4: Select Alex in JAWS
- Open the JAWS menu with Insert + J.
- Press Alt + O (Options) > V (Voices) > V (Voice Adjustment…).
- Under Synthesizer, select SAPI 5 (or SAPI 5×64).
- In the Voice Name drop-down list, select Alex (or any other Mac voice such as Victoria, Bruce, or Fred).
- Adjust your desired Rate, Pitch, and Punctuation level.
- Click Save As, name your profile (e.g., “Alex Voice”), and click OK.
Need Help With Setup or Facing Technical Glitches? We’ve Got You Covered!
Running into errors while configuring screen readers, installing speech engines, or fixing stubborn Windows software issues? Don’t stress—our expert assistive technology team is here to assist you.
Get quick, reliable, and friendly 1-on-1 remote tech support for your PC and assistive tools at very nominal and affordable rates.
Get Dedicated Tech Support Now
Troubleshooting Common Issues
The Synthesizer does not appear in NVDA / JAWS:
Make sure you copied all files inside the extracted folder (not the outer folder itself) into the speech data directory.
Access Denied / Cannot replace voice files:
Make sure your screen reader is actively using a different synthesizer before replacing the Alex.SpeechVoice folder.
Panthera SAPI doesn’t register voices:
Verify that the path %appdata%\macintalk-data contains the Speech folder directly inside it, then reopen panthera_settings.exe.
Download
download panthera_speech_nvda_addon
download panthera_sapi_setup file
Frequently Asked Questions (FAQs)
Which Mac OS X speech engine should I choose?
If you specifically want the Alex voice, choose Lion Speech (10.7) or Leopard Speech (10.5). If you want classic novelty voices (like Trinoids, Zarvox, or Whispering), Tiger Speech (10.4) or any of the other packs will work great.
Can I install multiple voice packs (e.g., Tiger and Lion) at the same time?
Yes. In NVDA, each engine has its own dedicated data folder under NVDA Menu > Tools > Mac OS X speech data, allowing you to switch between Tiger, Leopard, Snow Leopard, and Lion whenever you wish.
Does the Alex voice require an internet connection?
No. All speech processing is performed 100% offline on your local computer, ensuring zero latency and maximum responsiveness.
sútieto hlasy aj v inžch jazykoch ako angličtina? Ja by som mal záujem o slovenčinu.
Another update: On my HP laptop, ever since I installed the add-on and voices in NVDA, I get an error that says “Unable to create a timer. invalid Windows Handle”. What does this mean in plain English? I’m curious as to what I might have done wrong in the installation process while installing the add-on or the voices to get this error? How do I go about fixing it? Anyone else seeing this?
please update the addon to latest version. you can get it from the download section
Hey all! To get the Alex voice for sapi5, I’ve did everything as instructions say and registered the Lion voices in Panthera TTS. But, Jaws doesn’t recognize any of them. I’ve double checked the macintalk folder and all looks good there. Any advices how to get it up and running?
Just a quick update: I managed to get the Alex voice, both the Lion and the Sequoia versions, working in the NVDA add-on. I’m really liking it so far! I think the Sapi5 version for JAWS may be a bit of a tricky beast to tackle, but I’m certainly up for the challenge. 🙂 It’s really nice to be able to use Alex as well as the other Mac voices in Windows!
Just move the macintalk from the default appdata directory to another partition if you have, register voices in Panthera, and you can have Alex as a sapi5 voice!
I’m an average tech user who knows absolutely 0 when it comes to the developer side of things, so if anyone could lease help me understand (explain it to me in laymen’s terms), I would be grateful for any clarification you are able to provide.
I’m curious, would it not be more feasible for the voice data to be directly integrated into the NVDA add-on or the Sapi5 executable installer? I’m wondering because, having read through the instructions, this feels like an incredibly complex and convoluted installation/setup process for the average end user. I’d be curious to understand the limitations (if any), of integrating the voice data directly into the add-on or the executable, especially given that for some TTS add-ons, the vice data is already included, and all one must do is install the add-on, change the speech synthesizer, and voila. I’ve tried a similar add-on where I needed to install the add-on and install the voice data separately, and try as I might, reading the instructions and even seeking expert help, I was unsuccessful in getting the add-on to work, and this was on two different computers. So if there is a way to include the voice data into the add-on itself, I think it’s going to be a lot easier to install and set up, especially for the average end user like myself
Is there a way to get german voicepacks ?
How can someone boost the speech rate when using the Alex with jaws? when I’m increasing the speech rate when it’s reached 20 it’s count it as 100% and it’s not matching what I want.
alt ctrl windows page up
Thank you for this key shortcut. But I would like to ask that is it not possible to use the one registered with SAP5 with nvda instead of having it in two places on a computer? And I pray we have the text to speech on Android operating system as well.
the voice that you are using in nvda is through the addon, jaws don’t understand that so you need to use sapi5. currently there is no availability of this voice on android. we might work on the android alex text to speech, if we get enough views on this post. kindly suggesting you and others to share this post.
Could you please make audio tutorial for clear demonstration?