speech-alignment

Easy-to-use Speech Toolkit including Self-Supervised Learning model, SOTA/Streaming ASR with punctuation, Streaming TTS with text frontend, Speaker Verification System, End-to-End Speech Translation and Keyword Spotting. Won NAACL2022 Best Demo Award.

Cross-platform speech toolset, used from the command-line or as a Node.js library. Includes a variety of engines for speech synthesis, speech recognition, forced alignment, speech translation, voice isolation, language detection and more.

A robust forced alignment tool for low-resource languages using multiple ASR models and CER-based matching. Built for noisy data and imperfect transcripts.

Quran recitation alignment, segmentation and timing benchmark

Agentic video caption QA and subtitle rendering CLI: contextual ASR correction, word-level forced alignment, active-word ASS subtitles, and proof frames.

A cooperative project for the creation of an open-source corpus of spoken interactions in Romance.

Quran clip maker: turn any recitation into captioned short videos for TikTok, Reels and YouTube Shorts, with word-level timing. Runs in the browser. feedback to qc@okurdi.com

MMS Forced alignment package for JavaScript, TypeScript and Node.js: word-level timestamps from audio + transcript for captions, subtitles and karaoke. Meta MMS (wav2vec2) via ONNX Runtime with a CTC Viterbi decoder. Hindi, Hinglish and 20 Indic languages. No Python.

Native macOS & Apple Silicon offline bilingual audiobook compiler with MLX Whisper acoustic alignment and Apple Books typography.

Sync-first Korean subtitle (SRT) generator — Whisper + Korean wav2vec2 forced alignment for tight cue timing

Audio-video lip-sync quality and timing alignment analyzer measuring drift and sync scores in pure PHP

Dub a voice-over into another language with an AI agent: TTS timed to the original speaker, one profile per client

Word-level timestamps for Quranic recitation audio via DTW alignment transfer against reference recitations — no ASR.