SetMaxTokensPerSegment in go binding causes the segements to end prematurely · ggml-org/whisper.cpp#984

(1 comment) (1 reaction) (0 assignees)C++ (49,693 stars) (5,535 forks)batch import

buggood first issuehelp wanted

説明

When I use context.SetMaxTokensPerSegment and context.NextSegment, it generates incomplete segments. e.g. when tested using the jfk.wav sample, it gives the following output:

wspr_test.go:33: {Text:And so my fellow Americans ask Start:0s End:30s}

when this value is not set, it gives the correct output:

wspr_test.go:33: {Text:And so my fellow Americans ask not what your country can do for you Start:0s End:8s}
wspr_test.go:33: {Text:ask what you can do for your country. Start:8s End:11s}

test case: https://github.com/JackKCWong/go-whisper-test/blob/main/pkg/whisper/wspr_test.go#L29

コントリビューターガイド

技術スタック: gocpp
領域: backend
Issue 種別: bug
難度: 3
推定時間: 1-3 hours
活動状況: stale
明確さ: clear
前提条件: Gobasic understanding of whisper.cpp
初心者向け度: 50
調査方針: Look at the Go binding implementation, specifically the SetMaxTokensPerSegment and NextSegment functions in the whisper package. The test case in the issue provides a reproduction. Investigate how the max tokens per segment parameter is passed to the C++ code and whether there is an off by one or incorrect handling that causes segments to end prematurely. Compare with the behavior when the parameter is not set.