- 0.3–0.5s E2E — impossible w/ cloud solutions
- One-time hardware cost, zero per-call API fee
- Rock-steady — network never breaks flow
- Major languages supported with zero config
- Voice tiers: Machine / Simulated / Human
- ~10s sample audio suffices for voice cloning
- Text-only upload saves cloud bandwidth & cost
- Voice never leaves device — privacy compliant
- No cloud risk — regional, rate or sunset
- Core dialogue works offline & weak networks
Multi-lang ASR · Interpreting · Natural TTS
On-device multi-lang recognition, real-time translation & natural TTS. Deploy on desktop, meeting terminals, guide kiosks for two-way dialogue & cross-language communication.
Core Advantages
- Multilingual Recognition: Major languages ready out of the box
- Simultaneous Interpretation: Listen-and-translate w/ 0.3–0.5s E2E latency
- Tiered Voice Quality: Machine / Simulated / Human — pick by budget
Voice device control & field data entry
Voice replaces UIs & scanners in warehouses & workshops. Workers use natural language for logging, inspection, patrol forms & alerts. Local ASR outputs structured text for WMS, MES & IoT.
Core Advantages
- Lower Op Barrier: Natural language replaces UIs, scanners & work-order apps
- Weak-Network Ready: Local ASR w/ text backhaul — independent of on-site bandwidth
- Structured Output: Results feed directly into WMS, MES & work-order systems
Instant Wake · Local Control · Voiceprint
XIAO ESP32S3 low-power wake frontend triggers ASR-TTS on AI box. Voiceprint ID for per-member preferences. Integrates w/ Matter, HomeAssistant, Mi Home. Fully local — offline won't disrupt use.
Core Advantages
- Milliamp-Class Wake Frontend: ESP32S3 ESP-SR always-on, lasting months on battery
- Voiceprint: Identify members & load personal preferences
- Local Control: Integrated w/ Matter, HomeAssistant, Mi Home & more
Three Models: Frontend, Hybrid, All-in-One
Voice compute placement determines capability ceiling & per-unit BOM. Three common models:
Core Advantages
- Frontend (ESP32S3): Low-power wake & simple commands; pair w/ your host system
- Hybrid (Frontend + Box): Edge wakes + ASR + TTS; LLM remote. Best value.
- All-in-One (High-End): Single Jetson, full ASR/TTS/LLM. Privacy & offline.
Selection Guide| Product | Tier | Voice Capabilities | Demo Voice | Price |
|---|---|---|---|---|
 XIAO ESP32-S3 Sense | Wake Frontend | Wake Word / Command | — | ~$10 |
 reRouter CM4 | Entry | Single-lang ASR | — | $200–300 |
 reComputer AI R2130-12 | Entry | Multilingual Dialogue | Machine | ~$339 |
 reComputer J4012 | Professional | Dialogue + Cloning | Simulated | $800–900 |
 reComputer J5012 | Flagship | Dialogue + Clone + LLM | Natural | ~$2,000 |
Choose AI Compute Box by Capability
Compute boxes tiered by voice capabilities. Table lists tiers, capabilities, voice quality & price range. (See next tab for mic & speaker selection.)
Core Advantages
- Wake & Commands → Wake frontend, ~$10 all-in-one
- Dialogue → Mainstream tier; Natural TTS + cloning → Professional tier
- Voice + local LLM → Flagship tier, full pipeline on single device
| Product | Type | Use Case | Key Specs |
|---|---|---|---|
 ReSpeaker Lite | Pickup (Near-field) | ≤ 3m / Desktop / Workstation | 2-Mic / AI Audio / USB · I²S |
 ReSpeaker XVF3800 | Pickup (Mid/Far-field) | 3–5m / Meeting / Living Room | 4-Mic / XMOS DSP / AEC / Wake |
 ReSpeaker Flex Circular-4 | Pickup+Speaker (Circular) | Robot 360° / Wake Frontend | 4-Mic / XMOS DSP / AEC / 10W |
 ReSpeaker Flex Linear-4 | Pickup+Speaker (Linear) | Robot 180° / Wake Frontend | 4-Mic / XMOS DSP / AEC / 10W |
Pick Mics by Distance, Speakers by Shape
Key mic selection factors: pickup distance & ambient noise. Specs & recommended audio I/O combos below.
Core Advantages
- Pickup distance determines array size: ≤3 m use 2-Mic, 3–5 m use 4-Mic
- AEC: Mandatory when speaker & mic share enclosure. XVF3800 handles on DSP.
- Noise & Directionality: Noisy sites need hardware DSP — software insufficient
- Wake Frontend: XVF3800 kit w/ ESP32S3 for standalone wake-word — host can sleep