🎙️ High-precision, privacy-first speech intelligence platform powered by Faster-Whisper Turbo & SpeechBrain ECAPA-TDNN on CUDA. Features acoustic- first │ VAD segmentation, mutual k-NN graph Laplacian diarization, and multi-sample biometric voiceprint enrollment.
react postgresql cuda pytorch speech-to-text fastify speaker-diarization voice-biometrics bullmq ecapa-tdnn speechbrain pgvector faster-whisper speech-intelligence modal-com
-
Updated
Sep 2, 2026 - Python