murmure and murmurai

These are **competitors**: both offer self-hosted, GPU-accelerated speech-to-text solutions as alternatives to paid APIs, with murmure focusing on local privacy and LLM post-processing while murmurai emphasizes speaker diarization and API drop-in compatibility.

murmure
63
Established
murmurai
60
Established
Maintenance 23/25
Adoption 10/25
Maturity 13/25
Community 17/25
Maintenance 6/25
Adoption 11/25
Maturity 25/25
Community 18/25
Stars: 585
Forks: 56
Downloads:
Commits (30d): 45
Language: TypeScript
License: AGPL-3.0
Stars: 39
Forks: 16
Downloads: 67
Commits (30d): 0
Language: Python
License: MIT
No Package No Dependents
No risk flags

About murmure

Kieirra/murmure

Fully local, private and cross platform Speech-to-Text with LLM Post-processing

Leverages NVIDIA's Parakeet TDT 0.6B v3 neural network for low-latency on-device transcription across 25 European languages, with optional LLM post-processing for grammar correction and translation. Provides cross-platform desktop installers (Windows MSI/WinGet, Linux .deb/AppImage, macOS universal) and CLI configuration import for bulk deployment, enabling seamless voice dictation into AI editors like Cursor and ChatGPT without network connectivity.

About murmurai

namastexlabs/murmurai

🎙️ Drop-in replacement for paid transcription APIs. Self-hosted, GPU-powered, speaker diarization. Free forever: uvx murmurai

Wraps WhisperX in a REST API with word-level timestamps, multiple subtitle/JSON export formats, and webhook callbacks for async job completion. Built on FastAPI with SQLite persistence, GPU model caching, and background task processing via async workers. Provides SSRF-protected audio URL ingestion and integrates with pyannote for speaker diarization via HuggingFace tokens.

Scores updated daily from GitHub, PyPI, and npm data. How scores work