Fix tokens processing for byte-level BPE (#333)

k2-fsa · Sep 22, 2023 · 43b2b77 · 43b2b77
1 parent 969fff5
commit 43b2b77
Show file tree

Hide file tree

Showing 3 changed files with 40 additions and 15 deletions.
diff --git a/CMakeLists.txt b/CMakeLists.txt
@@ -1,7 +1,7 @@
 cmake_minimum_required(VERSION 3.13 FATAL_ERROR)
 project(sherpa-onnx)
 
-set(SHERPA_ONNX_VERSION "1.7.17")
+set(SHERPA_ONNX_VERSION "1.7.18")
 
 # Disable warning about
 #

diff --git a/sherpa-onnx/csrc/offline-stream.cc b/sherpa-onnx/csrc/offline-stream.cc
@@ -8,9 +8,9 @@
 
 #include <algorithm>
 #include <cmath>
+#include <iomanip>
 
 #include "kaldi-native-fbank/csrc/online-feature.h"
-#include "nlohmann/json.hpp"
 #include "sherpa-onnx/csrc/macros.h"
 #include "sherpa-onnx/csrc/offline-recognizer.h"
 #include "sherpa-onnx/csrc/resample.h"
@@ -256,25 +256,50 @@ const OfflineRecognitionResult &OfflineStream::GetResult() const {
   return impl_->GetResult();
 }
 std::string OfflineRecognitionResult::AsJsonString() const {
-  nlohmann::json j;
-  j["text"] = text;
-  j["tokens"] = tokens;
-#if 1
-  // This branch chooses number of decimal points to keep in
-  // the return json string
   std::ostringstream os;
-  os << "[";
+  os << "{";
+  os << "\"text\""
+     << ": ";
+  os << "\"" << text << "\""
+     << ", ";
+
+  os << "\""
+     << "timestamps"
+     << "\""
+     << ": ";
+  os << "\"[";
+
   std::string sep = "";
   for (auto t : timestamps) {
     os << sep << std::fixed << std::setprecision(2) << t;
     sep = ", ";
   }
+  os << "]\", ";
+
+  os << "\""
+     << "tokens"
+     << "\""
+     << ":";
+  os << "[";
+
+  sep = "";
+  auto oldFlags = os.flags();
+  for (const auto &t : tokens) {
+    if (t.size() == 1 && static_cast<uint8_t>(t[0]) > 0x7f) {
+      const uint8_t *p = reinterpret_cast<const uint8_t *>(t.c_str());
+      os << sep << "\""
+         << "<0x" << std::hex << std::uppercase << static_cast<uint32_t>(p[0])
+         << ">"
+         << "\"";
+      os.flags(oldFlags);
+    } else {
+      os << sep << "\"" << t << "\"";
+    }
+    sep = ", ";
+  }
   os << "]";
-  j["timestamps"] = os.str();
-#else
-  j["timestamps"] = timestamps;
-#endif
+  os << "}";
 
-  return j.dump();
+  return os.str();
 }
 }  // namespace sherpa_onnx
diff --git a/sherpa-onnx/csrc/symbol-table.cc b/sherpa-onnx/csrc/symbol-table.cc
@@ -51,7 +51,7 @@ void SymbolTable::Init(std::istream &is) {
     if (id >= 3 && id <= 258 && sym.size() == 6 && sym[0] == '<' &&
         sym[1] == '0' && sym[2] == 'x' && sym[5] == '>') {
       std::ostringstream os;
-      os << std::hex << (id - 3);
+      os << std::hex << std::uppercase << (id - 3);
 
       if (std::string(sym.data() + 3, sym.data() + 5) == os.str()) {
         uint8_t i = id - 3;