Quick recap for anyone new: Castwright turns a book into a full-cast audiobook — every character in its own voice, kept consistent across a series — rendered locally on your own machine. Source-available (FSL), solo project, in beta. Here’s what’s new in v1.12.
Captions and subtitles, from any finished book. Export .srt or .vtt — line, sentence, or word-level, whole-book or per-chapter — for read-along, accessibility, or cutting a demo/social clip. Line and sentence captions ride the timing Castwright already tracks, so there’s no extra render; word-level adds a single Whisper pass at export time. It runs through the same export queue as M4B/MP3.
The Pinokio one-click install now works end-to-end — and I want to be straight about it. The earlier one-click install had two bugs stacked on each other. v1.12.1 fixed the first (the installer couldn’t run at all), which immediately exposed the second: the server started from the wrong folder and quietly ignored its own settings — falling back to bare defaults instead of your workspace, worker counts, GPU budget and analyser config. v1.12.2 fixes that too. A clean Pinokio Install → Start now comes up fully configured from the first run. If you bounced off the install before, it’s sorted — please give it another go.
Casting reads a little truer. A rebuilt dialogue-attribution engine (deterministic, evidence-based, across all five languages) sharpens who-said-what — especially the tagless lines that used to fall to the wrong voice.
Where it’s still honest: solo project, beta; voice cloning (“your own voice”) is still in development; no OCR for scanned/image-only PDFs yet; and I’d still love testers reading in Spanish, French or German, and anyone on an AMD card (supported via ROCm, but I’ve no AMD hardware to verify it myself).
Hear a sample: https://www.castwright.ai/#demo · Full release notes: https://github.com/dudarenok-maker/Castwright/wiki/Release-Notes · castwright.ai · Feedback or a hand: hello@castwright.ai.
Quick recap for anyone new: Castwright turns a book into a full-cast audiobook — every character in its own voice, kept consistent across a series — rendered locally on your own machine. Source-available (FSL), solo project, in beta. Here’s what’s new in v1.12.
Captions and subtitles, from any finished book. Export .srt or .vtt — line, sentence, or word-level, whole-book or per-chapter — for read-along, accessibility, or cutting a demo/social clip. Line and sentence captions ride the timing Castwright already tracks, so there’s no extra render; word-level adds a single Whisper pass at export time. It runs through the same export queue as M4B/MP3.
The Pinokio one-click install now works end-to-end — and I want to be straight about it. The earlier one-click install had two bugs stacked on each other. v1.12.1 fixed the first (the installer couldn’t run at all), which immediately exposed the second: the server started from the wrong folder and quietly ignored its own settings — falling back to bare defaults instead of your workspace, worker counts, GPU budget and analyser config. v1.12.2 fixes that too. A clean Pinokio Install → Start now comes up fully configured from the first run. If you bounced off the install before, it’s sorted — please give it another go.
Casting reads a little truer. A rebuilt dialogue-attribution engine (deterministic, evidence-based, across all five languages) sharpens who-said-what — especially the tagless lines that used to fall to the wrong voice.
Where it’s still honest: solo project, beta; voice cloning (“your own voice”) is still in development; no OCR for scanned/image-only PDFs yet; and I’d still love testers reading in Spanish, French or German, and anyone on an AMD card (supported via ROCm, but I’ve no AMD hardware to verify it myself).
Hear a sample: https://www.castwright.ai/#demo · Full release notes: https://github.com/dudarenok-maker/Castwright/wiki/Release-Notes · castwright.ai · Feedback or a hand: hello@castwright.ai.