ggml-org/whisper.cpp

can I enable "-dl" option and detect multiple languages (in one audio-clip)? Thanks!

Open

#3,018 建立於 2025年4月8日

在 GitHub 查看
 (2 留言) (0 反應) (0 負責人)C++ (49,693 star) (5,535 fork)batch import
enhancementgood first issue

描述

Hi,

I could run whisper-cli to detect language and it works great (as the following): whisper-cli -f ./jfk.wav -dl --model ./ggml-small.bin

it works great and I could get the auto-detected language (like the following)::

...
whisper_full_with_state: auto-detected language: en (p = 0.958143)
...

my question is: can I also use the same "-dl" option to detect a wav-file which has multiple languages? in this case will whisper-cli show 2 language-types (similar as the above, but show 2 language-types)? Thanks!

貢獻者指南

技術棧
cpp
領域
climachine learning
議題類型
feature
難度面向新貢獻者的預計實作難度，1 表示很小改動，5 表示專家級工作。
3
預計時間有經驗貢獻者完成調查、實作、測試並準備 pull request 的粗略時間範圍。
over 1 week
活動狀態議題目前的可參與程度：新鮮、活躍、陳舊、阻塞或等待維護者輸入。
fresh
清晰度議題是否清楚說明預期改動、驗收標準和下一步。
clear
前置要求
C++whisper.cpp build process
新手友善度1-100 的估計分數，表示該議題對首次貢獻者的友善程度。
15
研究方向
The issue asks whether the dl flag can detect multiple languages in a single audio clip. To answer, investigate the current language detection implementation in whisper.cpp, specifically the `whisper full with state` function and how the dl flag triggers detection. Check if the model supports per segment language detection or only per clip. Consider examining the documentation and existing PRs or discussions about multi language support. A potential solution might involve modifying the CLI to output detected languages per segment or adding a new option.