Speaches
OpenAI API-compatible server for speech-to-text and text-to-speech
3.5k stars 427 forks last commit first released MIT
Actively maintained
Last commit 18 Apr 2026.
Speaches is an OpenAI API-compatible server for speech-to-text, translation, and text-to-speech, designed to be a local “model server” for voice workflows. It supports streaming and realtime interactions so applications can transcribe or generate audio with minimal integration changes.
Key Features
- OpenAI API compatibility for integrating with existing OpenAI SDKs and tools
- Streaming transcription via Server-Sent Events (SSE) for incremental results
- Speech-to-text powered by faster-whisper, with support for transcription and translation
- Text-to-speech using Piper and Kokoro models
- Realtime API support for low-latency voice interactions
- Dynamic model loading and offloading based on request parameters and inactivity
- CPU and GPU execution support
- Deployable with Docker and Docker Compose and designed to be highly configurable
Use Cases
- Replace hosted speech APIs with a self-managed, OpenAI-compatible voice backend
- Build realtime voice assistants that need streaming STT and fast TTS responses
- Batch transcription/translation pipelines for recordings with optional sentiment analysis
Speaches is a practical choice when you want OpenAI-style endpoints for voice features while retaining control over models and infrastructure. It fits well into existing OpenAI-oriented application stacks while focusing specifically on TTS/STT workloads.
Categories:
Tags:
Tech Stack:
Similar to Speaches
LocalAI
OpenAI-compatible local AI inference server and API
Run LLMs, image, and audio models locally with an OpenAI-compatible API, optional GPU acceleration, and a built-in web UI for managing and testing models.
Ollama
Run and manage large language models locally with an API
Ollama is a local LLM runtime that lets you pull, run, and customize models, offering a CLI and REST API for chat, generation, and embeddings.
Scriberr
Offline AI audio and video transcription with transcript chat
Scriberr is a self-hosted, privacy-focused AI transcription app for audio and video, with speaker diarization, word-level timestamps, summaries, and transcript chat.
AnythingLLM
All-in-one AI chat app with RAG, agents, and multi-model support
AnythingLLM is an all-in-one desktop and Docker app for chatting with documents using RAG, running AI agents, and connecting to local or hosted LLMs and vector databases.
Willow
Open-source, privacy-focused voice assistant platform
Self-hosted voice assistant platform for ESP32 devices with on-device wake-word and command recognition, Home Assistant integration, and an optional inference server for STT/TTS/LLM.
Khoj
Open-source personal AI for chat, semantic search and agents
Self-hostable personal AI 'second brain' for chat, semantic search, custom agents, automations and integration with local or cloud LLMs.