Merge branch 'ArmDeveloperEcosystem:main' into main

Author: zachlasiuk

.wordlist.txt

@@ -5638,3 +5638,10 @@ ZYNQ
 zypper
 ZZa
 ZZZZZ
+ASCT
+ArmTotalPerformance
+Numbat
+SKUs
+asct
+geminicli
+passwordless

content/install-guides/asct.md

@@ -92,7 +92,7 @@ wget https://artifacts.tools.arm.com/asct/dist/0.4.2/asct-0.4.2+3b955c2.tar.gz
 
 ### How do I install ASCT using uv?
 
-The recommended installation method uses [uv](https://github.com/astral-sh/uv), a fast Python package installer. First, install `uv`:
+The recommended method uses [uv](https://github.com/astral-sh/uv), a fast Python package installer. Install `uv`:
 
 ```bash
 curl -LsSf https://astral.sh/uv/install.sh | sh
@@ -108,7 +108,7 @@ This installs ASCT to `/usr/local/bin` making it available system-wide. Installi
 
 ### How do I verify that ASCT is installed?
 
-After installing ASCT, verify the installation by checking the version:
+Verify the installation by checking the version:
 
 ```bash
 asct version
@@ -120,19 +120,19 @@ The output is similar to:
 ASCT 0.4.2+3b955c2
 ```
 
-You can also display the help information:
+Display the help information:
 
 ```bash
 asct --help
 ```
 
-The output displays available commands and benchmarks.
+This displays available commands and benchmarks.
 
 ## How do I use ASCT?
 
 ASCT provides several commands for benchmarking and system analysis, including `run`, `system-info`, `list`, `diff`, and `sysreg`. 
 
-Some benchmarks require `sudo` or root privileges to configure huge pages and access certain system information. You can run ASCT without `sudo`, but some benchmarks might be unavailable or limited in functionality.
+Some benchmarks require `sudo` or root privileges to configure huge pages and access certain system information. You can run ASCT without `sudo`, but some benchmarks can be unavailable or limited in functionality.
 
 ### Get system information
 
@@ -269,7 +269,7 @@ Some of the tests also generate graphs.
 
 An example of a bandwidth graph is shown below:
 
-![Example bandwidth sweep benchmark output showing memory bandwidth measurements across different data sizes on an Arm system, with a graph displaying bandwidth in GB/s on the y-axis and data size in MB on the x-axis](/install-guides/_images/asct-bw.webp)
+![Graph showing memory bandwidth sweep benchmark results with data size in MB on x-axis ranging from 0.001 to 1000, and bandwidth in GB/s on y-axis ranging from 0 to 300, displaying multiple colored lines representing different CPU configurations with performance peaks at smaller data sizes and declining bandwidth as data size increases beyond cache levels alt-txt#center](/install-guides/_images/asct-bw.webp "Bandwidth sweep benchmark results")
 
 ## How do I uninstall ASCT?
 

content/install-guides/atp.md

@@ -1,9 +1,6 @@
 ---
-### Title the install tools article with the name of the tool to be installed
-### Include vendor name where appropriate
 title: Arm Total Performance
 
-### Optional additional search terms (one per line) to assist in finding the article
 additional_search_terms:
 - performance
 - profiling
@@ -13,20 +10,18 @@ additional_search_terms:
 - perf
 - top-down
 
-### Estimated completion time in minutes (please use integer multiple of 5)
 minutes_to_complete: 30
 
+author: Pareena Verma
+
 ### Link to official documentation
 official_docs: https://arm-total-performance.tools.arm.com/
 
-author: Pareena Verma
-
-### PAGE SETUP
-weight: 1                       # Defines page ordering. Must be 1 for first (or only) page.
-tool_install: false             # Set to true to be listed in main selection page, else false
-multi_install: false            # Set to true if first page of multi-page article, else false
-multitool_install_part: false   # Set to true if a sub-page of a multi-page article, else false
-layout: installtoolsall         # DO NOT MODIFY. Always true for tool install articles
+weight: 1                  
+tool_install: false        
+multi_install: false        
+multitool_install_part: false
+layout: installtoolsall
 
 build:
   list: false                   # Exclude from .Pages collections and listings
@@ -140,7 +135,7 @@ Verify the installation by checking the version:
 ./atperf version
 ```
 
-The output displays the installed version number.
+The output shows the installed version number:
 
 ```output
 Arm Total Performance CLI version: 0.39.0
@@ -176,7 +171,7 @@ Before connecting to an Arm Linux target, ensure SSH key-based authentication is
 
 ### Check your existing SSH key
 
-Verify that you can connect to your target using SSH:
+Verify that you can connect to your target using SSH. Replace `user` with your username and `target_host` with your target's hostname or IP address:
 
 ```bash
 ssh user@target_host
@@ -190,7 +185,7 @@ ssh-keygen -y -f ~/.ssh/id_ed25519
 
 If you're prompted for a passphrase, your key is protected with one. ATP doesn't support SSH keys with passphrases, so you need to create a separate key without a passphrase for ATP.
 
-### Create a passphrase-free key for ATP (if needed)
+## Create a passphrase-free key for ATP (if needed)
 
 If your existing SSH key has a passphrase, generate a new key specifically for ATP:
 
@@ -212,9 +207,9 @@ If your existing key is in the default location (`~/.ssh/id_ed25519` or `~/.ssh/
 ssh-copy-id -i ~/.ssh/atp_key.pub user@target_host
 ```
 
-You can now use `~/.ssh/atp_key` as the private key when configuring ATP's target connection.
+You can now use `~/.ssh/atp_key` as the private key when configuring the ATP target connection.
 
-### Enable passwordless sudo on the target
+## Enable passwordless sudo on the target
 
 ATP needs to run commands with elevated privileges on the target system. Because ATP cannot enter sudo passwords interactively, configure passwordless sudo access.
 
@@ -238,7 +233,7 @@ After installing ATP and preparing your target, you can connect using either the
 
 ### Connect using the ATP GUI
 
-The GUI is typically used on Windows and macOS hosts, though it's also available on Linux.
+Use the GUI on Windows and macOS hosts, though you can also run it on Linux.
 
 Open the ATP application.
 
@@ -248,24 +243,25 @@ Select **Add Target**.
 
 In the **Configure Target** form, provide the following details:
 
-* **Host**: The hostname or IP address of the target machine
-* **Name**: A descriptive name for the target
-* **Port**: The SSH port number (default is 22)
-* **User**: The username for SSH connection
-* **SSH Private Key**: Choose **Automatically Detect Key** or **Select Key Manually**
+* **Host**: the hostname or IP address of the target machine
+* **Name**: a descriptive name for the target
+* **Port**: the SSH port number (default is 22)
+* **User**: the username for SSH connection
+* **SSH Private Key**: choose **Automatically Detect Key** or **Select Key Manually**
   * For manual selection, enter the path to your private key (usually `~/.ssh/id_rsa` or `~/.ssh/id_ed25519`)
 * **Host Key Verification**: Choose **Strict** (recommended) or **Ignore**
   * Strict mode verifies the server identity using your `~/.ssh/known_hosts` file
 
-![ATP target configuration example #center](_images/atp-target-config.png)
+![ATP target configuration form displaying input fields including Host with IP address 192.168.1.10, Name field for descriptive target identifier, Port field set to 22, User field for SSH username, SSH Private Key section with radio buttons for Automatically Detect Key and Select Key Manually options, and Host Key Verification dropdown menu set to Strict mode with explanation text about verifying server identity using known_hosts file alt-txt#center](/install-guides/_images/atp-target-config.png "Configure Target form with connection settings")
+
 
 Select **Add Target**.
 
 The target appears in the targets list and is ready for profiling.
 
 You can select the **Test Connection** button to verify your connection to the ATP Linux target is successful.
 
-![ATP connection test #center](_images/atp-connection-test.png)
+![ATP graphical interface displaying connection test results with green checkmark icon indicating success, target name my-target shown in header, host IP address 192.168.1.10 listed below, connection status field showing Connected in green text, and blue Test Connection button at bottom of panel#center](/install-guides/_images/atp-connection-test.png "Successful ATP target connection test")
 
 #### Configure jump nodes (optional)
 
@@ -310,7 +306,7 @@ atperf --help
 
 If your target is the same machine where you installed ATP (the host machine itself), ATP automatically provides a built-in `localhost` target. You don't need to configure authentication, SSH keys, or host verification for local profiling.
 
-Verify the localhost target is available:
+Verify the localhost target is available by listing all configured targets:
 
 ```bash
 atperf target list
@@ -328,7 +324,7 @@ Your local machine is now ready to use as a target. Specify `--target=localhost`
 
 #### Add a remote target
 
-To profile a workload on a remote machine, add it as a target using the standard syntax:
+To profile a workload on a remote machine, add it as a target using the standard syntax. This command registers the remote system with ATP:
 
 ```bash
 atperf target add <user>@<host>:<port>:<private_ssh_key_path>
@@ -452,11 +448,11 @@ atperf target remove my-target
 
 To remove ATP from your system, use the appropriate method for your platform:
 
-**Windows**: Open **Settings** > **Apps**, find **Arm Total Performance**, and select **Uninstall**.
+* For Windows, open **Settings** > **Apps**, find **Arm Total Performance**, and select **Uninstall**.
 
-**macOS**: Drag the Arm Total Performance application from **Applications** to the **Trash**.
+* For macOS, drag the Arm Total Performance application from **Applications** to the **Trash**.
 
-**Linux**: Remove the package:
+* For Linux, remove the package:
 
 ```bash
 sudo apt remove arm-total-performance

content/learning-paths/cross-platform/_example-learning-path/questions.md

@@ -6,12 +6,6 @@ weight: 11
 layout: "learningpathall"
 ---
 
-### Why is the search box missing on the home page?
-
-The search index is not automatically generated, but you can add it to enable the search box. 
-
-Refer to the information about the search box in the [Learning Path setup](/learning-paths/cross-platform/_example-learning-path/setup/#search) section.
-
 ### Why are my Learning Path pages in the wrong order?
 
 Each markdown file has a value called weight.

content/learning-paths/laptops-and-desktops/dgx_spark_voicechatbot/1_offline_voice_assistant.md

@@ -1,35 +1,31 @@
 ---
-title: Understanding Offline Voice Assistants
+title: Learn about offline voice assistants
 weight: 2
 
 ### FIXED, DO NOT MODIFY
 layout: learningpathall
 ---
 
-## Why Build an Offline Voice Assistant?
+## Why build an offline voice assistant?
 
 Voice-based AI assistants are becoming essential in customer support, productivity tools, and embedded interfaces. For example, a retail kiosk might need to answer product-related questions verbally without relying on internet access. However, many of these systems depend heavily on cloud services for speech recognition and language understanding, raising concerns around latency, cost, and data privacy.
 
 In addition, a healthcare terminal or legal consultation assistant may need to handle voice queries involving sensitive personal information, where sending audio data to the cloud would violate privacy requirements. Running your voice assistant entirely offline solves these problems. 
 
 You avoid unpredictable latency caused by network fluctuations, prevent sensitive voice data from leaving the local machine, and eliminate recurring API costs that make large-scale deployment expensive. It also boosts trust for on-device deployments and compliance-sensitive industries.
 
-By combining local speech-to-text (STT) with a locally hosted large language model (LLM), you gain full control over the pipeline and eliminate API dependencies. This gives you full control to experiment, customize, and scale—without relying on external APIs.
+By combining local speech-to-text (STT) with a locally hosted large language model (LLM), you gain complete control over the pipeline and eliminate API dependencies. You can experiment, customize, and scale without relying on external services.
 
-## Common Development Challenges:
+## What are some common development challenges?
 
-While the benefits are clear, building a local voice assistant involves several engineering challenges:
+While the benefits are clear, building a local voice assistant involves several engineering challenges.
 
-- Managing audio stream segmentation and speech detection in real-time: It's hard to reliably identify when the user starts and stops speaking, especially with natural pauses and background noise.
-
-- Avoiding latency or misfires in STT/LLM integration: Timing mismatches can cause delayed responses or repeated input, reducing the conversational quality.
-
-- Keeping the pipeline responsive on local hardware without overloading resources: You need to carefully balance CPU/GPU workloads so that inference doesn't block audio capture or processing.
+Real-time audio segmentation requires reliably identifying when users start and stop speaking, accounting for natural pauses and background noise. Timing mismatches between STT and LLM components can cause delayed responses or repeated input, reducing conversational quality. You also need to balance CPU/GPU workloads to keep the pipeline responsive without overloading resources or blocking audio capture.
 
 ## Why use Arm and DGX Spark?
 
-Arm-powered platforms like [DGX Spark](https://www.nvidia.com/en-gb/products/workstations/dgx-spark/) allow efficient parallelism: use CPU cores for audio preprocessing and whisper inference, while offloading LLM reasoning to powerful GPUs. This architecture balances throughput and energy efficiency—ideal for private, on-prem AI workloads. Check this [learning path](https://learn.arm.com/learning-paths/laptops-and-desktops/dgx_spark_llamacpp/1_gb10_introduction/) to understand the CPU and GPU architecture of DGX Spark.
+Arm-powered platforms like [DGX Spark](https://www.nvidia.com/en-gb/products/workstations/dgx-spark/) allow efficient parallelism: use CPU cores for audio preprocessing and whisper inference, while offloading LLM reasoning to powerful GPUs. This architecture balances throughput and energy efficiency—ideal for private, on-premises AI workloads. To understand the CPU and GPU architecture of DGX Spark, refer to [Unlock quantized LLM performance on Arm-based NVIDIA DGX Spark](/learning-paths/laptops-and-desktops/dgx_spark_llamacpp/).
 
-DGX Spark also supports standard USB interfaces, making it easy to connect consumer-grade microphones for development or deployment. This makes it viable not just for data center use, but also for edge inference or desktop-style prototyping.
+DGX Spark also supports standard USB interfaces, making it easy to connect consumer-grade microphones for development or deployment. This makes it viable for edge inference and desktop-style prototyping.
 
 In this Learning Path, you’ll build a complete, offline voice chatbot prototype using PyAudio, faster-whisper, and vLLM on an Arm-based system—resulting in a fully functional assistant that runs entirely on local hardware with no internet dependency.