Skip to content
@VoxRT

VoxRT

On-device lightweight voice layer including TTS, ASR, VAD, wake-word and more. Cross-platform: Android, iOS, embedded Linux.

VoxRT

Voice AI that runs on device. Runtime and models built from scratch, for the CPU.

We build two halves of the same product:

  1. A custom inference runtime in Rust, written for streaming audio on commodity CPUs: no GPU, no NPU, no vendor accelerator in the critical path. One code base runs as NEON on ARMv8 (phones, Raspberry Pi, Jetson, Graviton), AVX2 on x86-64 and SIMD128 in WebAssembly.
  2. Audio models that run on it: text-to-speech, streaming speech recognition, wake word, keyword spotting and voice activity detection.

Real deployments do not have a spare GPU. They have one ARM core, a battery indicator the user is watching, and a privacy policy that says audio never leaves the device. On a $15 Raspberry Pi Zero 2 W our wake word takes 5 % of one A53 core, sustained; on a phone our text-to-speech starts speaking about 40 ms after it gets the text.

What we ship

Product Android iOS Linux Browser Models
LinnetTTS, text-to-speech (41 M, English, streaming every 80 ms, voice prompts) LinnetTTS (AAR) same repo (Swift Package) same repo (C, Python, Rust) demo on voxrt.com linnet-en-35m, demo Space
Wake word ("Hey Assistant", ~100 KB model) voxrt-wake-word-android voxrt-wake-word-ios voxrt-wake-word-linux (aarch64) voxrt-wake-word-browser voxrt-wake-word-models
Keyword spotting (14 commands, 636 K-parameter streaming Conformer) voxrt-kws-android voxrt-kws-ios voxrt-kws-linux (x86-64, aarch64) voxrt-kws-browser voxrt-kws-models
Streaming ASR (NeMo FastConformer 32 M, CTC + RNN-T, cache-aware streaming) voxrt-asr-android voxrt-asr-ios voxrt-asr-linux (x86-64, aarch64) demo on voxrt.com voxrt-asr-models
Voice activity detection (Silero v5) voxrt-silero-android voxrt-silero-ios on request voxrt-silero-models

Channels are the ones developers already use: JitPack and Gradle on Android, Swift Package Manager on iOS, PyPI, npm, go get, C tarballs with CMake and pkg-config, Rust as a git dependency, npm for the browser packages. The Linux SDKs are one .so per architecture on a glibc 2.17 baseline, so the same binary runs on Raspberry Pi 3, 4, 5 and Zero 2, NVIDIA Jetson, AWS Graviton and any x86-64 server from Haswell on. The models are mirrored on Hugging Face.

All four engines run side by side in a browser tab at voxrt.com/try.

Numbers

Measured on the shipped builds, one thread unless stated. RTF = compute time / audio time, lower is better.

Product Device Result
LinnetTTS Apple M5 3.3 ms per 80 ms of speech (RTF 4 %)
LinnetTTS iPhone 13 Pro Max (A15) 5.5 ms per step (RTF 7 %), first audio 38 ms after the text
LinnetTTS Snapdragon 662, 4 threads 14 ms per step (RTF 18 %)
LinnetTTS Raspberry Pi Zero 2 W, 4 threads 59 ms per step (RTF 74 %), still real time
LinnetTTS Core i9-13900K, one core RTF 10.6 % including the prime pass, first audio 22 ms, 187 MB peak
Wake word Raspberry Pi Zero 2 W (A53 @ 1 GHz) RTF 5.3 %
Wake word Snapdragon 662 (A73) / iPhone 13 Pro Max RTF 2.1 % / 1.5 %
Wake word, WebAssembly iPhone 13 Pro Max, Safari RTF 0.23 %
Keyword spotting Snapdragon 662 (A73) / iPhone 13 Pro Max RTF 16 % / 8.9 %
Keyword spotting x86-64 AVX2 / Raspberry Pi Zero 2 W RTF 1 % / 32 %
Streaming ASR x86-64 AVX2 RTF 5.9 % (CTC), 8.0 % (RNN-T); WER 3.27 % (RNN-T) on a 500-utterance LibriSpeech test-clean subset
Streaming ASR iPhone 13 Pro Max / Snapdragon 662, live microphone RTF 8–10 % / 35 %
Voice activity detection Snapdragon 662 / iPhone 13 Pro Max RTF 3.05 % / 1.85 %

Every repository has the method behind its numbers (device, clocks, chunk sizes, warm-up) and the scripts to reproduce them. LinnetTTS also has a side-by-side benchmark against Piper, Kokoro and pocket-tts on the same CPU, with audio.

The runtime

Typical mobile / edge runtimes VoxRT runtime
Binary size 5–20 MB about 600 KB per engine
Streaming audio retrofitted the only mode
ARMv8 kernels partial universal, one binary from Cortex-A53 to Apple M-series
x86-64 / browser separate stacks the same kernels as AVX2 + FMA and WebAssembly SIMD128
Hot-path allocations many none (pre-allocated scratch)
Numerics "looks about right" bit-exact across scalar, NEON, AVX2 and WebAssembly
Model weights at rest plain files encrypted .vxrt for the native voice SDKs; plain safetensors for LinnetTTS
Scalar vs NEON undocumented 8.7× on Cortex-A73, method published

Adding a model is wiring its weights into the existing operator set, not a new deployment story. The same team wrote the models: the wake word, keyword spotting and LinnetTTS models are trained in house; the ASR is NVIDIA's FastConformer (CC-BY-4.0) and the VAD is Silero v5 (MIT), both ported to the runtime bit-exactly.

Licensing

  • Voice SDKs (wake word, keyword spotting, ASR, VAD): the Kotlin, Swift, Python, Node.js, Go and C wrappers are Apache-2.0 in the open; the compiled runtime is a proprietary binary that is free to use and redistribute inside your product. Wake-word and keyword-spotting weights are ours and free to use with the SDK; the ASR weights are CC-BY-4.0, the VAD weights MIT.
  • LinnetTTS: the whole runtime, bindings and tools are Apache-2.0 source. The weights are under the Linnet Weights License: free in your own products, commercial included, with the built-in voices; custom voices, fine-tuning, training on the outputs and general-purpose TTS services need a separate licence.

What we sell

Models tuned to your product on the same runtime: wake phrases with your brand name and your languages, keyword sets, speech-to-intent, domain ASR, custom TTS voices. The open libraries above are the proof that the runtime ships; the custom work is what funds it.

Licensing, OEM integration, custom model packaging: help@voxrt.com · voxrt.com

How we work

  • CPU first. A single ARMv8 core is the target. Accelerator paths are a return-on-investment question, not the foundation.
  • One binary across the CPU tier. Universal NEON kernels, same code on a cheap A53 and a flagship X-series core; runtime feature detection is opt-in, not load-bearing.
  • Battery-aware by construction. No allocations on the hot path, no f64 where f32 works, every kernel profiled against its audio budget before it ships.
  • Bit-exact or it does not land. NEON, AVX2 and WebAssembly must match the scalar reference within the ULP budget; the TTS reproduces the same audio for the same seed on a given platform at any thread count.
  • Numbers with a method. Every published figure names the device, the clock, the chunk size and the warm-up, and most come with a script.

Targets

  • iOS (iPhone, iPad, arm64), macOS (Apple silicon, for LinnetTTS)
  • Android (arm64-v8a, x86_64 emulator)
  • Linux x86-64 (Haswell and later) and aarch64 (glibc 2.17+): Raspberry Pi 3 / 4 / 5 / Zero 2, NVIDIA Jetson, AWS Graviton, Rock Pi / Orange Pi / Khadas boards
  • Browser (WebAssembly SIMD128, no server)
  • Microcontrollers and other targets on request

Stack

Rust · ARMv8 NEON and x86-64 AVX2 intrinsics · WebAssembly SIMD128 · cbindgen · pyo3 / maturin · napi-rs · cgo · wasm-bindgen · cargo-ndk · Swift Package Manager · JitPack · PyPI · npm · Go modules · Xcode xcframework · GitHub Actions


If you are putting voice into a product and the CPU budget, the battery or the privacy policy is the constraint, we want to talk.

Popular repositories Loading

  1. LinnetTTS LinnetTTS Public

    Lightweight on-device text-to-speech: flow LM over a continuous-latent codec, Rust runtime, Android/iOS/Linux/Python

    Rust 37

  2. voxrt-asr-android voxrt-asr-android Public

    Streaming on-device speech recognition for Android — NEON-accelerated FastConformer (32M params), cache-aware streaming with 80 ms look-ahead, no cloud. Powered by the VoxRT runtime.

    Kotlin 21 1

  3. voxrt-silero-android voxrt-silero-android Public

    On device streaming voice activity detection (Silero VAD v5) for Android. ~424 KB native binary, NEON-accelerated arm64-v8a, RTF ~3% on Snapdragon 662.

    Kotlin 17 4

  4. voxrt-asr-ios voxrt-asr-ios Public

    Streaming on-device speech recognition for iOS — NEON-accelerated, encrypted FastConformer (32M params), RTF 0.08–0.10 on iPhone 13 Pro Max. Built on the VoxRT custom Rust inference runtime. SwiftP…

    Swift 16 1

  5. voxrt-silero-ios voxrt-silero-ios Public

    On device streaming voice activity detection (Silero VAD v5) for iOS. Custom Rust inference runtime, NEON-accelerated arm64, RTF ~1.85% on iPhone.

    Swift 14 1

  6. voxrt-asr-models voxrt-asr-models Public

    Pre-compiled ASR model weights for the VoxRT on-device runtime. Encrypted .vxrt v2 format. streaming-medium-pc: FastConformer 32M, CTC + RNN-T, CC-BY-4.0 (NVIDIA NeMo).

    13

Repositories

Showing 10 of 19 repositories

Top languages

Loading…

Most used topics

Loading…