What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Build the companion as two connected systems: a speech-and-AI pipeline that reports what it is doing, and a Rive character that turns those states and speech into an animated face. The basic flow is microphone → speech recognition → AI response → text-to-speech → application events and audio data → Rive runtime → display. The face can react while the assistant is listening or waiting, rather than appearing only after a response is ready.
What you need for a first build
A voice-enabled screen companion needs a Raspberry Pi, a display, a microphone, a speaker, and software for the conversation pipeline and the animated face. A Raspberry Pi 5 is a practical reference platform, but the cited hardware example is not a requirement for every design.
- Required for the visible voice prototype: Raspberry Pi, display, microphone, and speaker.
- Optional: touchscreen, camera, push-to-talk or stop buttons, enclosure, and battery or UPS.
- Not required for Rive animation: an AI accelerator. The accelerator matters only if your chosen AI workload benefits from local inference hardware.
A reference build uses a Raspberry Pi 5 with 8 GB or 16 GB of memory, a 5-inch DSI display, USB microphone, and USB or amplified speaker. Treat those as example components, not a universal compatibility list. An HDMI screen can also work, and a headless setup is useful while developing the voice pipeline.
Choose where speech recognition and AI run
Decide early whether the assistant will process requests locally, depend on cloud services, or combine both. This choice affects internet and account requirements, model and speech capabilities, and setup and maintenance—not how Rive animates the face.
#1 Best Overall
- Includes Raspberry Pi 5 with 2.4Ghz 64-bit quad-core CPU (8GB RAM)
- Includes 128GB Micro SD Card pre-loaded with 64-bit Raspberry Pi OS, USB MicroSD Card Reader
- CanaKit Turbine Black Case for the Raspberry Pi 5
- CanaKit Low Noise Bearing System Fan
- Mega Heat Sink - Black Anodized
| Approach | Documented example | What to consider |
|---|---|---|
| Local pipeline | A community project documents openWakeWord, energy-based voice activity detection, whisper.cpp, Ollama with gemma3:1b and optional moondream vision, Piper TTS, and a pygame face. Its reference hardware includes a Raspberry Pi 5, display, microphone, and speaker. Community project |
This is a concrete software example, not a benchmark or a Rive implementation. Your model, workload, and setup determine performance. |
| Cloud-backed agent | ElevenLabs documents a Raspberry Pi voice-agent path that requires a Pi 5 or similar, microphone, speaker, Python 3.9 or later, an account, and an API key. ElevenLabs Raspberry Pi guide | This introduces a service and account dependency. Check the applicable service terms and policies before building around it. |
The available examples do not establish a controlled comparison of local and cloud latency, recognition accuracy, privacy, or cost. Evaluate those on your own hardware and with your chosen service, workload, and usage pattern; do not assume local processing is automatically private, fast, or free.
Define the interface between the assistant and the face
Keep conversation logic separate from animation details. The application should send a small, stable set of events to Rive, and the Rive file should decide how those events look. That lets you revise expressions and transitions without rewriting microphone or AI code.
Rank #2
- CanaKit Raspberry Pi 5 Essentials Starter Kit
A useful example contract has one activity mode and one independent emotion value. For instance, the mode could use 0 for idle, 1 for listening, 2 for thinking, 3 for speaking, 4 for connecting, 5 for error, and 6 for sleeping. Emotion could use named values such as neutral, happy, empathetic, concerned, or surprised. These are suggested values, not required input names or a mandatory Rive schema; choose and document your own types and names.
- Listening: acknowledge capture with focused eyes or a visible listening indicator.
- Thinking: use subtle eye movement or a small pulse so the character does not freeze while the request is processed.
- Speaking: animate the mouth from audio data while allowing the emotion to remain independently controlled.
- Connecting and error: make connection waits and failures visibly different from thinking; send the appropriate state when a request fails or connectivity drops.
- Idle and sleeping: use restrained blinking or small pupil motion. Constant movement can distract.
Plan the transitions as well as the values: stop speaking when playback is interrupted, return to listening when a new capture starts, and move to an error or connecting state when the request lifecycle requires it. Praneeth Kawya Thathsara, a Rive animator and interactive character specialist at Mascot Engine, puts the design challenge this way: “A convincing AI companion needs more than a voice.” Read the article
Rank #3
- Pi5 8GB Pack: RasTech Pi 5 8GB kit includes 1 x Pi5 8GB board ,1 x 64GB Card, 2 x Card Readers,1 x Active Cooler,1 x Case for Pi5, 2 x 4K Micro HD Out Cable,1 x GaN 27W 5A USB-C Power supply,1 x Screwdriver and 1 x instructions.
- Pi5 8GB Board: The Pi5 board is equipped with a 64-bit quad-core Arm Cortex-A76 processor running at 2.4GHz and an 800MHz VideoCore VII GPU with support for OpenGL ES 3.1 and Vulkan 1.2, which delivers a significant increase in graphics performance. Dual HD Out 4Kp60 display outputs and a built-in dual 4-channel MIPI camera/display transceiver provide state-of-the-art camera support. The Pi 5 offers a 2-3 times increase in CPU performance compare to Pi4.
- Important Graphics Features: Equipped with an 800MHz VideoCore VII GPU and providing better graphics performance, suitable for multimedia applications,gaming,and graphics intensive tasks.Provides 1 UART interface,1 card slot that supports high-speed operation, 2 USB. 3 0.5 ports that support synchronous 0Gbps operation,2 USB 2.0 port ports,2 4Kp60 display outputs that support HDR.Built-in dedicated dual 4-channel 1Gbps MIPI DSI/CSI connectors,triple the total bandwidth.
- Cooling Kit for Pi 5: Compatible with Active Cooler for Raspberry Pi5, It can provide Pi 5 board with better cooling effect in using. The Case can accurately access usb-c power jack,Micro HD Out ports, usb ports, Ethernet jack, card slot, power button, 4-lane MIPI DSI/CSI connectors and so on, and it also supports installation of cooling fan.
- 64GB Card Kit and GaN 27W USB-C Power Supply: With extra 64GB card to store more files and card readers for multiple medium, keep better performance for Raspberry Pi 5, 27W USB C Power Supply is Compatible with Pi5 8GB, offers a variety of output voltage options, including 5.1V at 5A, 9.0V at 3.0A, 12.0V at 2.25A, and 15.0V at 1.8A, providing for different device requirements.
Build the speech and AI loop
- Choose an input method. Add a microphone for voice interaction. A camera is optional for visual questions or camera-based gaze; buttons can provide push-to-talk, stop, or volume controls. Touch or a cursor can also steer gaze without a camera.
- Capture speech and mark the state. When capture begins, send the listening state to the character. On capture end, stop showing active listening and start the recognition and response stages.
- Recognize speech and request a response. Run the speech-recognition and assistant components locally, through a cloud service, or across both, according to your chosen architecture. Send thinking or connecting events during waits rather than leaving the face idle without feedback.
- Generate and play speech. When text-to-speech audio is ready, send the speaking state and audio data to the display application. When playback ends or is interrupted, update the activity state promptly.
- Handle failures explicitly. If recognition, the assistant request, audio playback, or connectivity fails, move to an error or connecting state and provide a clear route back to listening.
For a local-pipeline reference, the community repository describes wake-word detection, voice activity detection, speech recognition, an LLM, optional vision captioning, and TTS. It uses a pygame face, so it illustrates one possible assistant pipeline rather than proving a Rive integration. View the community project
Design the Rive character and integrate its runtime
Create the character and its state machine in Rive, with inputs matching the event contract your application will send. Connect those inputs to the face’s listening, thinking, speaking, connecting, error, idle, and optional sleeping behaviors. Keep facial emotion separate from activity where possible; for example, a speaking character can also look concerned.
Rank #4
- A RASPBERRY PI 5 KIT FROM AN APPROVED RESELLER: This Vilros Complete Starter Kit for Pi 5 Includes Raspberry Pi 5 Board with all the accessories you need to get started.
- 9 PART KIT INCLUDES MOST ACCESSORIES NEEDED YOU TO GET UP AND RUNNING: 1. Raspberry Pi 5 Board–2.Metal/Aluminum Alloy Passive & Active Cooling Case–3.Raspberry Pi 5 Compatible Power Supply–4. PWM fan With 10k Max RPM Capacity (pre-installed in the case)--5. 32GB Micro SD Card With 64bit Raspberry Pi OS Preinstalled–6. Standard HDMI to Micro HDMI Adapter Cable--7.Neoprene Storage bag–8.Vilros Quickstart Guide for Raspberry Pi–9. Mini To Standard Camera Module Adapter Cable to use a camera module with a PI 5
- RASPBERRY PI 5 SPECS AND FEATURES:--Processor: Broadcom BCM2712 2.4GHz quad-core 64-bit Arm Cortex-A76 CPU, with cryptography extensions, 512KB per-core L2 caches, and a 2MB shared L3 cache----Features: 2.4GHz quad-core, 64-bit Arm Cortex-A76 CPU–VideoCore VII GPU supporting Vulkan 1.2 and OpenGL ES–LPDDR4X-4267 SDRAM (4GB and 8GB options)--PCIe 2.0 x1 interface for fast peripherals ( Requires adapter)--Dual-band 802.11ac Wi-Fi 2.4 GHz and 5.0 GHz –Bluetooth 5.0 / Bluetooth Low Energy (BLE)
- MULTIFUNCTION PASSIVE & ACTIVE COOLED CASE: The case features a built-in pole/column that contacts the main chip on the Raspberry Pi 5 board via an included thermal pad to passively cool the board and also includes a preinstalled PWM Fan that plugs directly into the fan port on the board. The fan will only turn on if needed and will also increase RPMs as needed. Other features include a built-in power button that shows the onboard light status, camera module compatibility, and can be used in the single-layer configuration for hat compatibility
- HIGH-QUALITY COMPONENTS: All components are manufactured with Raspberry Pi in mind and are backed by the Vilros 1-Year warranty.
Rive documents runtime options including Web and C++. That does not mean every runtime works the same way on every Raspberry Pi installation. The suitable implementation depends on the Pi’s operating system, graphics stack, application framework, and display setup. Check the runtime and rendering backend against your actual Linux environment before committing to an integration. Rive runtime documentation
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Clear out junk files and repair common Windows errors3Scan for outdated or missing drivers - takes under a minute- Build the state machine. Create clearly named, typed inputs for activity and any independent emotion or gaze controls.
- Choose the rendering route. Select a Rive runtime that fits your application framework and supported graphics environment; verify the combination on the target Pi.
- Connect application events. Update Rive inputs when recording starts or stops, the request progresses, speech begins or ends, or an error occurs.
- Test on the intended screen. Check the display resolution and rendering behavior on the actual LCD, touchscreen, or HDMI display, not only in the editor.
Make the mouth follow speech
For a lightweight mouth animation, measure the current TTS audio amplitude and map it to a normalized input such as 0–1, then use that value to control mouth opening. The range is an illustrative control scale, not a measured synchronization result.
Best Value
- Includes Raspberry Pi 5 16GB with 2.4Ghz 64-bit quad-core CPU (16GB RAM)
- Includes 128GB Micro SD Card pre-loaded with 64-bit Raspberry Pi OS, USB MicroSD Card Reader
- CanaKit Turbine Black Case for the Raspberry Pi 5
- CanaKit Low Noise Bearing System Fan
- Mega Heat Sink - Black Anodized
For more expressive articulation, send viseme or phoneme-derived mouth-shape values if your chosen TTS stack exposes them. Another option is to combine a mouth shape from viseme data with amplitude for movement intensity. These are implementation techniques; synchronization quality depends on the audio pipeline and how the application supplies timing and values.
Add optional gaze, vision, and local acceleration
Gaze can be driven by a camera, touch input, cursor, device orientation, or another sensor. A deliberately randomized idle gaze is another option. Keep idle behavior subtle so the face remains present without competing for attention.
A camera is useful only if you want camera-based gaze or visual questions. Likewise, a local AI accelerator is optional and does not render the Rive face. Raspberry Pi’s current Hailo documentation requires a Raspberry Pi 5 and 64-bit Raspberry Pi OS for its AI accelerator setup. The official guidance describes AI HAT+ for vision and moderate neural workloads, and AI HAT+ 2 for supported local LLM and VLM workloads; it says the earlier AI Kit is no longer in production and recommends the HAT+ models for new designs. Raspberry Pi AI documentation
The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →| Product | Vendor-stated specifications | Role in this project |
|---|---|---|
| AI HAT+ | 13 TOPS or 26 TOPS, depending on variant | Optional acceleration for supported workloads; not needed just to animate a Rive face. |
| AI HAT+ 2 | 40 TOPS, 8 GB onboard memory, documented for LLM/VLM workloads up to approximately 6 billion parameters | Optional for supported local LLM/VLM use. These are vendor specifications, not measured performance for this companion. |
If installing an AI HAT, follow the current assembly and software instructions and check supported package versions. Raspberry Pi recommends powering the Pi down before installation and recommends an Active Cooler; for AI HAT+ 2 it recommends both the Pi Active Cooler and the HAT’s supplied heatsink. AI HAT assembly guidance
Quick Recap
Bring-up checklist and common failure points
- No visible face: verify that the selected runtime and graphics backend support the Pi’s operating system and application framework, then check display output independently.
- Face remains frozen during waits: send listening, thinking, connecting, and error events from the request lifecycle instead of updating the character only when a response arrives.
- Mouth does not match audio: confirm the application is sampling the audio that is actually playing and updating the Rive input over time; amplitude-based motion is a simple approximation, while viseme data can provide more specific shapes when available.
- Assistant does not return to the right state: define transitions for playback completion, interruption, request failure, and lost connectivity.
- Local models exceed the intended setup: check the model and workload against the selected Pi and supported accelerator configuration rather than assuming an AI HAT is required or that its TOPS figure predicts application performance.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
