Metadata-Version: 2.4
Name: oremi-andika
Version: 3.0.0b11
Summary: Empowering seamless transcription with cutting-edge STT (Speech-to-Text) technology, revolutionizing interaction through accurate speech recognition
Project-URL: Say Thanks!, https://www.buymeacoffee.com/demsking
Project-URL: Documentation, https://demsking.gitlab.io/oremi-andika
Project-URL: Source, https://gitlab.com/demsking/oremi-andika
Project-URL: Issues, https://gitlab.com/demsking/oremi-andika/-/issues
Author-email: Sébastien Demanou <demsking@gmail.com>
Maintainer-email: Sébastien Demanou <demsking@gmail.com>
License-Expression: Apache-2.0
License-File: LICENSE
License-File: LICENSE_HEADER.txt
Keywords: stt
Classifier: Development Status :: 2 - Pre-Alpha
Classifier: Environment :: Console
Classifier: Intended Audience :: Developers
Classifier: Natural Language :: English
Classifier: Natural Language :: French
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Requires-Python: >=3.12
Requires-Dist: numpy>=2.2.6
Requires-Dist: openapi-pydantic>=0.5.1
Requires-Dist: pydantic-settings>=2.14.2
Requires-Dist: pydantic>=2.10.0
Requires-Dist: qwen-asr[vllm]>=0.0.6
Requires-Dist: safetensors>=0.8.0
Requires-Dist: soundfile>=0.14.0
Requires-Dist: torch>=2.0.0
Requires-Dist: torchaudio>=2.9.1
Requires-Dist: vosk>=0.3.45
Description-Content-Type: text/markdown

# Oremi Andika

[![License](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](LICENSE)
[![Support](https://img.shields.io/badge/Support-Buy%20me%20a%20beer-orange.svg)](https://www.buymeacoffee.com/demsking)

**Oremi Andika** serves as the foundational pillar of the Oremi project,
introducing cutting-edge STT (Speech-to-Text) capabilities to enable seamless
transcription of spoken language across multiple engines.

Built upon a unified speech recognition architecture, this
module empowers the Oremi personal assistant to accurately transcribe spoken words
into text using multiple backends (Vosk and Mega-ASR), facilitating effortless
communication and interaction.

Drawing inspiration from the Swahili word for _'transcribe'_, Andika
encapsulates the essence of transforming spoken language into written text. By
embodying this linguistic concept, Oremi Andika not only facilitates efficient
communication but also bridges the gap between spoken and written expression,
empowering users to seamlessly interact with their personal assistant through
natural speech while enjoying the convenience of textual representation.

![Oremi Andika Logo](htdocs/logo.png)

## Features

- **Real-Time Streaming**: Stream audio bytes directly to the server and receive
  recognition results incrementally with boundary overlap deduplication.
- **Extensible STT Platform**: Clean abstraction separating application
  logic from specific speech recognition backends.
- **Multiple STT Backends**: Native support for **Vosk** and **Mega-ASR**.
- **Audio Quality Routing & LoRA Switching**: Dynamic routing of noisy/degraded
  audio using an audio quality router and zero-latency LoRA delta switching.
- **vLLM Acceleration**: Ultra-fast KV-cache streaming inference with
  materialized LoRA checkpoints.
- **Privacy-First**: Operates completely offline without external cloud API
  dependencies.
- **Docker Support**: Ready-to-use Docker images for easy orchestration on CPU
  or GPU.
- **Home Assistant Integration**: Standard-compliant integration with Home
  Assistant's Assist voice pipeline.

## Documentation

Full documentation, including setup guides, API reference, and integrations,
is available at: **[demsking.gitlab.io/oremi-andika](https://demsking.gitlab.io/oremi-andika/)**

## License

Copyright 2023-2026 Sébastien Demanou. All Rights Reserved.

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    https://gitlab.com/demsking/oremi-andika/-/blob/main/LICENSE

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
