Oct 26 Today 00:40 v0.3.0: TensorFold 1.1.5; What's new in v0.3.0; builds and Docker tag 0.3.0 Oct 25 Yesterday 22:49 v0.2.2: DGX Spark Full voice ~0.5 s voice to voice 20:23 v0.2.1: What's new (NVRTC fix for fresh DGX Spark installs, no "(Boo says)"), Mac Drama 3 first audio ~60 ms, image 0.2.1 drowzeys +1: Claude Opus 5.5 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PfjmufXm4L7qsVNFyRw9ZK 19:54 v0.2.0: DGX Spark install connection fix in the fixes list 19:16 v0.2.0: What's new (Kokoro Lite how-to, Lite vs Drama 3 table, fixes) near the top; Docker 0.2.0 Lite installs for DGX Spark and Mac, Docker VOICE_ENGINE=lite / both, the Full/Lite toggle, trade-offs, the Kokoro Lite vs Drama 3 (4-bit) table and the fixes since v0.1.0. The DGX Spark Drama 3 build now has TensorFold 1.1.4's silence guard flags. 17:51 Full vs Lite requirements table and one-shot Lite recipes (DGX Spark, Mac, Docker) 17:45 Lite voice (Kokoro-82M): requirements, measured comparison, how to choose, install switch, credits; five Kokoro previews Card: optional lite voice section and requirements; Credits table adds Kokoro-82M, kokoro, mlx-audio, misaki, spaCy, espeak-ng and StyleTTS 2. previews/kokoro/: synthetic Kokoro-82M stock voices (bf_emma, bf_isabella, bm_george, af_heart, am_michael). 13:37 Card: Drama 3's silence guard for long lines that go quiet (TensorFold 1.1.4) 02:03 Card: Install section with the one-shot installer 01:23 Card: the 46-second demo right under the title 01:22 Card: Silero credit points to the GitHub repo 01:22 Card: powered-by-TensorFold title, promo video, screenshots, TensorFold 1.1.4 notes, Credits table Oct 24 2 days ago 23:45 Fourteen synthetic Drama 3 voices for Boo, with previews 
