Skip to content
@soniqo

Soniqo

Realtime conversational voice AI.

Soniqo

Speech and audio apps, open-source SDKs, and ML engineering services for your products and private infrastructure.

Website · ML Engineering Services · Documentation · Soniqo on LinkedIn

Mimi — in development

Meet Mimi → — A small robot with planned local voice control through a nearby laptop or phone. Assembly and voice-control development are in progress.

Open-source SDKs

speech-swift — Speech recognition, synthesis, speech-to-speech, VAD, diarization and audio processing for Apple Silicon, using MLX and CoreML.

speech-android — Kotlin SDK and Android integration for local speech recognition, synthesis, VAD and audio processing.

speech-core — C++ speech runtime and voice-agent pipelines, with ONNX Runtime and LiteRT backends.

Model downloads: Apple models · ONNX and LiteRT models

Apps

App What it does
Soniqo Music Free vocal removal, stem separation, key and speed changes, and WAV export on Android, iPhone and iPad.
speech CLI Transcription, speech synthesis and audio tools in your terminal on an Apple Silicon Mac. Install with brew install speech.
Speech Studio Local voice cloning and scripted dialogue with multiple speech engines. Available on macOS, Windows and Linux.
Stenograf Record, transcribe and search conversations on Mac, Windows and Android, with optional desktop summaries and chat.

ML Engineering Services

We create datasets, train and adapt models, and integrate them into your product. Work starts with your target tasks, data, languages and hardware.

  • Custom datasets — Collect, curate and annotate data for your task, languages and domain, with quality checks and separate training and evaluation sets.
  • Model training and adaptation — Adapt speech, audio and search models to your users, vocabulary and workflows.
  • Evaluation and benchmarking — Measure accuracy, quality, latency and memory on representative data and target hardware.
  • Inference and deployment — Convert and optimize models, integrate runtimes and SDKs, and deploy on devices or private infrastructure.
  • Audio and search engineering — Speech enhancement, spatial audio, retrieval, ranking and research workflows.

Discuss your project →

Documentation

Website languages: English · 中文 · 日本語 · 한국어 · Español · Deutsch · Français · हिन्दी · Português · Русский · العربية · Tiếng Việt · Türkçe · ไทย

Community and contact

Discord — Questions, support, model requests and updates.

Soniqo on LinkedIn · Blog · YouTube robots and demos

For integration support or model requests, reach out to Ivan →.

Pinned Loading

  1. speech-swift speech-swift Public

    AI speech toolkit for Apple Silicon — ASR, TTS, speech-to-speech, VAD, and diarization powered by MLX and CoreML

    Swift 1.3k 193

  2. speech-android speech-android Public

    On-device speech SDK for Android — ASR, TTS, VAD, and noise cancellation powered by ONNX Runtime with Qualcomm NNAPI acceleration

    Kotlin 179 13

  3. speech-core speech-core Public

    On-device VAD / streaming STT / TTS / diarization in C++17 (ONNX + LiteRT) with a voice-agent pipeline. Linux, Windows, Android.

    C++ 101 8

  4. speech-studio speech-studio Public

    Open-source desktop voice-cloning studio for creators — clone a voice, script lines with emotion markers, synthesize on-device. Tauri + VoxCPM2, runs on macOS, Windows, and Linux.

    TypeScript 87 15

Repositories

Showing 7 of 7 repositories
  • speech-swift Public

    AI speech toolkit for Apple Silicon — ASR, TTS, speech-to-speech, VAD, and diarization powered by MLX and CoreML

    soniqo/speech-swift's past year of commit activity
    Swift 1,284 Apache-2.0 193 8 7 Updated Nov 22, 2026
  • speech-studio Public

    Open-source desktop voice-cloning studio for creators — clone a voice, script lines with emotion markers, synthesize on-device. Tauri + VoxCPM2, runs on macOS, Windows, and Linux.

    soniqo/speech-studio's past year of commit activity
    TypeScript 87 Apache-2.0 15 1 0 Updated Nov 15, 2026
  • .github Public
    soniqo/.github's past year of commit activity
    0 0 0 0 Updated Nov 13, 2026
  • speech-android Public

    On-device speech SDK for Android — ASR, TTS, VAD, and noise cancellation powered by ONNX Runtime with Qualcomm NNAPI acceleration

    soniqo/speech-android's past year of commit activity
    Kotlin 179 Apache-2.0 13 1 0 Updated Nov 10, 2026
  • speech-core Public

    On-device VAD / streaming STT / TTS / diarization in C++17 (ONNX + LiteRT) with a voice-agent pipeline. Linux, Windows, Android.

    soniqo/speech-core's past year of commit activity
    C++ 101 Apache-2.0 8 1 0 Updated Nov 10, 2026
  • voice-tau2-bench Public

    Standalone voice-agent benchmark with local speech-swift inference on Apple Silicon.

    soniqo/voice-tau2-bench's past year of commit activity
    Python 1 MIT 0 0 0 Updated Oct 30, 2026
  • homebrew-tap Public

    Homebrew tap for Soniqo on-device speech tools

    soniqo/homebrew-tap's past year of commit activity
    Ruby 0 Apache-2.0 0 0 0 Updated Oct 25, 2026