diff --git a/history.md b/history.md index 560d283..3895ff0 100644 --- a/history.md +++ b/history.md @@ -52,3 +52,10 @@ The project now features a high-performance, Apple Silicon-optimized pipeline th - **Solution:** - Artificially truncated input transcription to a maximum of 250 characters. - Added `max_new_tokens=150` to the translation generation call to ensure the model terminates even if it gets stuck in a loop. + +## Phase 7: Multilingual Detection & Bridge Translation +- **Goal:** Support input in any language, detect it, and translate to English + others. +- **Approach:** + - Switched to `whisper-small-mlx` (multilingual). + - **Hub-and-Spoke Model:** If a non-English language is detected, Whisper's `task="translate"` is used to create an English "bridge" text, which is then fed into the specialized MarianMT models. +- **Outcome:** Full support for multilingual input with centralized translation.