From 94de39dde2e616a01683c518023d0fab72b4e103 Mon Sep 17 00:00:00 2001
From: 游雁 <zhifu.gzf@alibaba-inc.com>
Date: 星期一, 19 二月 2024 22:21:50 +0800
Subject: [PATCH] aishell example
---
runtime/onnxruntime/src/util.cpp | 34 +++++++++++++++++++++-------------
1 files changed, 21 insertions(+), 13 deletions(-)
diff --git a/runtime/onnxruntime/src/util.cpp b/runtime/onnxruntime/src/util.cpp
index ac793f5..a12570b 100644
--- a/runtime/onnxruntime/src/util.cpp
+++ b/runtime/onnxruntime/src/util.cpp
@@ -305,6 +305,10 @@
}
bool TimestampIsPunctuation(U16CHAR_T &u16) {
+ // (& ' -) in the dict
+ if (u16 == 0x26 || u16 == 0x27 || u16 == 0x2D){
+ return false;
+ }
return (u16 >= 0x21 && u16 <= 0x2F) // 鏍囧噯ASCII鏍囩偣
|| (u16 >= 0x3A && u16 <= 0x40) // 鏍囧噯ASCII鏍囩偣
|| (u16 >= 0x5B && u16 <= 0x60) // 鏍囧噯ASCII鏍囩偣
@@ -361,9 +365,13 @@
}
}
-std::string VectorToString(const std::vector<std::vector<int>>& vec) {
+std::string VectorToString(const std::vector<std::vector<int>>& vec, bool out_empty) {
if(vec.size() == 0){
- return "";
+ if(out_empty){
+ return "";
+ }else{
+ return "[]";
+ }
}
std::ostringstream out;
out << "[";
@@ -580,19 +588,18 @@
}
}
// format
- ts_sent += "{'text':'" + text_seg + "',";
- ts_sent += "'start':'" + to_string(start) + "',";
- ts_sent += "'end':'" + to_string(end) + "',";
- ts_sent += "'ts_list':" + VectorToString(ts_seg) + "}";
+ ts_sent += "{\"text_seg\":\"" + text_seg + "\",";
+ ts_sent += "\"punc\":\"" + characters[idx_str] + "\",";
+ ts_sent += "\"start\":" + to_string(start) + ",";
+ ts_sent += "\"end\":" + to_string(end) + ",";
+ ts_sent += "\"ts_list\":" + VectorToString(ts_seg, false) + "}";
if (idx_str == characters.size()-1){
ts_sentences += ts_sent;
} else{
ts_sentences += ts_sent + ",";
}
-
// clear
- idx_str++;
text_seg = "";
ts_sent = "";
start = 0;
@@ -605,9 +612,9 @@
text_seg += " " + characters[idx_str];
}
ts_seg.push_back(timestamps[idx_ts]);
- idx_str++;
idx_ts++;
}
+ idx_str++;
}
// for none punc results
if(ts_seg.size() >0){
@@ -618,10 +625,11 @@
end = ts_seg[ts_seg.size()-1][1];
}
// format
- ts_sent += "{'text':'" + text_seg + "',";
- ts_sent += "'start':'" + to_string(start) + "',";
- ts_sent += "'end':'" + to_string(end) + "',";
- ts_sent += "'ts_list':" + VectorToString(ts_seg) + "}";
+ ts_sent += "{\"text_seg\":\"" + text_seg + "\",";
+ ts_sent += "\"punc\":\"\",";
+ ts_sent += "\"start\":" + to_string(start) + ",";
+ ts_sent += "\"end\":" + to_string(end) + ",";
+ ts_sent += "\"ts_list\":" + VectorToString(ts_seg, false) + "}";
ts_sentences += ts_sent;
}
--
Gitblit v1.9.1