Add function 'tolowerUnicode' in sherpa-onnx-microphone (fix #791) (#812)

sherpa-onnx/csrc/sherpa-onnx-microphone.cc

@@ -7,7 +7,8 @@
 #include <stdlib.h>
 
 #include <algorithm>
-#include <cctype>  // std::tolower
+#include <clocale>
+#include <cwctype>
 
 #include "portaudio.h"  // NOLINT
 #include "sherpa-onnx/csrc/display.h"
@@ -37,6 +38,31 @@ static void Handler(int32_t sig) {
   fprintf(stderr, "\nCaught Ctrl + C. Exiting...\n");
 }
 
+static std::string tolowerUnicode(const std::string& input_str) {
+    
+    // Use system locale
+    std::setlocale(LC_ALL, ""); 
+    
+    // From char string to wchar string
+    std::wstring input_wstr(input_str.size()+1, '\0');
+    std::mbstowcs(&input_wstr[0], input_str.c_str(), input_str.size()); 
+    std::wstring lowercase_wstr;
+
+    for (wchar_t wc : input_wstr) {
+        if (std::iswupper(wc)) {
+            lowercase_wstr += std::towlower(wc);
+        } else {
+            lowercase_wstr += wc;
+        }
+    }
+    
+    // Back to char string
+    std::string lowercase_str(input_str.size()+1, '\0');
+    std:wcstombs(&lowercase_str[0], lowercase_wstr.c_str(), lowercase_wstr.size());
+
+    return lowercase_str;
+}
+
 int32_t main(int32_t argc, char *argv[]) {
   signal(SIGINT, Handler);
 
@@ -172,11 +198,7 @@ for a list of pre-trained models to download.
 
     if (!text.empty() && last_text != text) {
       last_text = text;
-
-      std::transform(text.begin(), text.end(), text.begin(),
-                     [](auto c) { return std::tolower(c); });
-
-      display.Print(segment_index, text);
+      display.Print(segment_index, tolowerUnicode(text));
       fflush(stderr);
     }