v1.15.0 v2.15.0

Name: speech-to-speech
Author: huggingface

huggingface/speech-to-speechAug 17, 2026 → Aug 17, 20260 days between releases

Generating AI comparison

Analyzing what changed between v1.15.0 and v2.15.0. This page updates automatically — usually within a minute.

The Two Releases

v1.15.0

1.15.0 — Selective OCR for scanned and mixed PDFs

Aug 17, 2026· 2 months ago

Introduces selective OCR capabilities for scanned and mixed PDFs using PDFium rendering and PP-OCRv6, enabling text layer recovery and adaptive fusion.

Full release notes

v2.15.0

Aug 17, 2026· 2 months ago

This release focuses on diagram layout improvements and AI model integration. It introduces a DeepSeek Harness bundle and fixes packaging issues on Windows, along with better handling of diagram elements like lanes and overlapping rectangles.

Full release notes