# Hangry Labs

> Hangry Labs builds local-first, private-by-default AI and open software for speech, transcription, LLM operations, automation, and creative work.

The website is localized, but the unprefixed URLs below are the canonical English pages. Product pages describe the supported workflow; linked project repositories contain README, license, and release documentation. Some desktop apps are proprietary; a public download repository does not imply open-source application code.

## Core pages

- [Hangry Labs](https://hangrylabs.app/hangry-labs): Project overview, design principles, and guided technology picker.
- [Docker containers](https://hangrylabs.app/software): Browse containerized projects and installation guides.
- [Desktop apps](https://hangrylabs.app/software/apps): Apps with official installer downloads and independent product pages.
- [GitHub organization](https://github.com/hangry-labs): Source repositories and release history.
- [About](https://hangrylabs.app/about): Background and public contact information.

## Desktop apps

- [Listen2Me](https://listen2me.hangrylabs.app/): Free-to-use, proprietary Windows x64 dictation app in beta; local recognition or a user-selected hosted endpoint.
- [Listen2Me releases](https://github.com/hangry-labs/Listen2Me/releases): Official standard Windows installer, release notes and SHA-256 checksums. Models are downloaded separately. Installers are self-signed.
- [Listen2Me feedback privacy](https://hangrylabs.app/listen2me/feedback-privacy): Website handling of optional feedback; the app's own privacy notice is linked from its product page.

## SSML-H

- [SSML-H 1.0 namespace](https://hangrylabs.app/ns/ssml-h/1.0): Canonical rendered overview for the Hangry Labs extension profile for W3C SSML 1.1.
- [SSML-H 1.0 specification](https://hangrylabs.app/ns/ssml-h/1.0/specification.md): Normative specification in Markdown.
- [SSML-H specification source](https://github.com/hangry-labs/ssml-h-spec): Specification source, proposal process, notices, and license.
- [SSML-H tools](https://github.com/hangry-labs/ssml-h-tools): Reference Python parser, validator, builder, and processor integration guidance.

## Speech and transcription projects

- [KokoroTTS](https://hangrylabs.app/software/kokorotts): Fast multilingual local text-to-speech project overview.
- [Install KokoroTTS](https://hangrylabs.app/software/kokorotts/install): Docker installation guidance.
- [KokoroTTS source](https://github.com/hangry-labs/kokoroTTS): README, API, benchmarks, examples, and releases.
- [MeloTTS](https://hangrylabs.app/software/melotts): Multilingual local speech studio overview.
- [Install MeloTTS](https://hangrylabs.app/software/melotts/install): Docker installation guidance.
- [MeloTTS source](https://github.com/hangry-labs/MeloTTS): README, API, examples, and releases.
- [OmniVoiceTTS](https://hangrylabs.app/software/omnivoicetts): Multilingual voice design, cloning, and SSML-H studio overview.
- [Install OmniVoiceTTS](https://hangrylabs.app/software/omnivoicetts/install): Docker installation guidance.
- [OmniVoiceTTS source](https://github.com/hangry-labs/OmniVoiceTTS): README, API, benchmarks, examples, and releases.
- [Faster Qwen3-TTS Studio](https://hangrylabs.app/software/faster-qwen3-tts-studio): Local voice design, cloning, dialogue, streaming, and API workspace overview.
- [Install Faster Qwen3-TTS Studio](https://hangrylabs.app/software/faster-qwen3-tts-studio/install): Docker installation guidance for the full and tiny images.
- [Faster Qwen3-TTS Studio source](https://github.com/hangry-labs/faster-qwen3-tts-studio): README, API, benchmark, model, license, and development documentation.
- [VoxCPMTTS](https://hangrylabs.app/software/voxcpmtts): Creative speech, voice design, and cloning workspace overview.
- [Install VoxCPMTTS](https://hangrylabs.app/software/voxcpmtts/install): Docker installation guidance.
- [VoxCPMTTS source](https://github.com/hangry-labs/VoxCPMTTS): README, API, benchmarks, examples, and releases.
- [Qwen3-ASR-STT](https://hangrylabs.app/software/qwen3-asr-stt): Local multilingual speech-to-text service overview.
- [Install Qwen3-ASR-STT](https://hangrylabs.app/software/qwen3-asr-stt/install): Docker installation guidance.
- [Qwen3-ASR-STT source](https://github.com/hangry-labs/Qwen3-ASR-STT): README, API, benchmarks, examples, MCP tools, and releases.

## Other projects

- [YuE2](https://hangrylabs.app/software/yue2): Local lyrics-to-song studio with external GGUF models under noncommercial terms.
- [Install YuE2](https://hangrylabs.app/software/yue2/install): NVIDIA Docker setup, pinned model downloads, and read-only mounting.
- [YuE2 source](https://github.com/hangry-labs/yue2): README, native API, model provenance, benchmark, and listening examples.
- [Sidon](https://hangrylabs.app/software/sidon): Local speech restoration, two-speaker separation, and audio conversion studio.
- [Install Sidon](https://hangrylabs.app/software/sidon/install): Full/tiny Docker images, CPU/NVIDIA modes, and immutable stable references.
- [Sidon source](https://github.com/hangry-labs/Sidon): README, native audio APIs, licenses, and before/after examples.

- [llama.nodrama](https://github.com/hangry-labs/llama.nodrama): Diagnostics for llama.cpp slots, KV cache behavior, and local inference operations.
- [MDelight](https://md.hangrylabs.app/): Markdown editor for web and Windows with live preview, encrypted documents, and reviewable AI suggestions.

## Optional

- [Suggestions](https://hangrylabs.app/suggestions): Curated browser and privacy tools.
- [Security policy](https://hangrylabs.app/.well-known/security.txt): Private vulnerability reporting contact and policy metadata.