extern/whisper.cpp
1
0
Fork 1
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-04-15 23:08:21 +02:00
Code Issues Packages Projects Releases Wiki Activity

whisper : only trim if split_on_word is true (#476)

Browse Source
This commit is contained in:
boolemancer 2023-02-07 22:43:23 -08:00 committed by GitHub
parent ab1916fc59
commit 4dd7119deb
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 8 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
whisper.cpp
View File

@ -3014,8 +3014,7 @@ static inline void trim(std::string &s) {
static inline bool should_split_on_word(const char * txt, bool split_on_word) { static inline bool should_split_on_word(const char * txt, bool split_on_word) {
if (!split_on_word) return true; if (!split_on_word) return true;
std::string s = txt; return txt[0] == ' ';
return s.substr(0, 1) == " ";
} }
// wrap the last segment to max_len characters // wrap the last segment to max_len characters
@ -3039,7 +3038,10 @@ static int whisper_wrap_segment(struct whisper_context & ctx, int max_len, bool
if (acc + cur > max_len && i > 0 && should_split_on_word(txt, split_on_word)) { if (acc + cur > max_len && i > 0 && should_split_on_word(txt, split_on_word)) {
// split here // split here
trim(text); if (split_on_word) {
trim(text);
}
ctx.result_all.back().text = std::move(text); ctx.result_all.back().text = std::move(text);
ctx.result_all.back().t1 = token.t0; ctx.result_all.back().t1 = token.t0;
ctx.result_all.back().tokens.resize(i); ctx.result_all.back().tokens.resize(i);
@ -3067,7 +3069,9 @@ static int whisper_wrap_segment(struct whisper_context & ctx, int max_len, bool
} }
} }
trim(text); if (split_on_word) {
trim(text);
}
ctx.result_all.back().text = std::move(text); ctx.result_all.back().text = std::move(text);
return res; return res;