Support Chinese heteronyms on Android for TTS. (#742)

Author: csukuangfj

cmake/openfst.cmake

@@ -3,18 +3,18 @@
 function(download_openfst)
   include(FetchContent)
 
-  set(openfst_URL  "https://github.com/kkm000/openfst/archive/refs/tags/win/1.6.5.1.tar.gz")
-  set(openfst_URL2 "https://hub.nuaa.cf/kkm000/openfst/archive/refs/tags/win/1.6.5.1.tar.gz")
-  set(openfst_HASH "SHA256=02c49b559c3976a536876063369efc0e41ab374be1035918036474343877046e")
+  set(openfst_URL  "https://github.com/csukuangfj/openfst/archive/792965fda2a3bc29f282321f527af0d6ba26fd22.zip")
+  set(openfst_URL2 "https://hub.nuaa.cf/csukuangfj/openfst/archive/792965fda2a3bc29f282321f527af0d6ba26fd22.zip")
+  set(openfst_HASH "SHA256=815d8acf555e4aaece294d6280ec209d0e9d91e0120e8406b24ff7124ecdbb26")
 
   # If you don't have access to the Internet,
   # please pre-download it
   set(possible_file_locations
-    $ENV{HOME}/Downloads/openfst-win-1.6.5.1.tar.gz
-    ${CMAKE_SOURCE_DIR}/openfst-win-1.6.5.1.tar.gz
-    ${CMAKE_BINARY_DIR}/openfst-win-1.6.5.1.tar.gz
-    /tmp/openfst-win-1.6.5.1.tar.gz
-    /star-fj/fangjun/download/github/openfst-win-1.6.5.1.tar.gz
+    $ENV{HOME}/Downloads/openfst-792965fda2a3bc29f282321f527af0d6ba26fd22.zip
+    ${CMAKE_SOURCE_DIR}/openfst-792965fda2a3bc29f282321f527af0d6ba26fd22.zip
+    ${CMAKE_BINARY_DIR}/openfst-792965fda2a3bc29f282321f527af0d6ba26fd22.zip
+    /tmp/openfst-792965fda2a3bc29f282321f527af0d6ba26fd22.zip
+    /star-fj/fangjun/download/github/openfst-792965fda2a3bc29f282321f527af0d6ba26fd22.zip
   )
 
   foreach(f IN LISTS possible_file_locations)

scripts/dotnet/generate.py

@@ -40,8 +40,8 @@ def process_linux(s):
         "libpiper_phonemize.so.1",
         "libsherpa-onnx-c-api.so",
         "libsherpa-onnx-core.so",
-        "libsherpa-onnx-fstfar.so.7",
-        "libsherpa-onnx-fst.so.6",
+        "libsherpa-onnx-fstfar.so.16",
+        "libsherpa-onnx-fst.so.16",
         "libsherpa-onnx-kaldifst-core.so",
         "libucd.so",
     ]
@@ -69,8 +69,8 @@ def process_macos(s):
         "libpiper_phonemize.1.dylib",
         "libsherpa-onnx-c-api.dylib",
         "libsherpa-onnx-core.dylib",
-        "libsherpa-onnx-fstfar.7.dylib",
-        "libsherpa-onnx-fst.6.dylib",
+        "libsherpa-onnx-fstfar.16.dylib",
+        "libsherpa-onnx-fst.16.dylib",
         "libsherpa-onnx-kaldifst-core.dylib",
         "libucd.dylib",
     ]

sherpa-onnx/csrc/offline-tts-vits-impl.h

@@ -56,6 +56,9 @@ class OfflineTtsVitsImpl : public OfflineTtsImpl {
       }
       std::vector<std::string> files;
       SplitStringToVector(config.rule_fars, ",", false, &files);
+
+      tn_list_.reserve(files.size() + tn_list_.size());
+
       for (const auto &f : files) {
         if (config.model.debug) {
           SHERPA_ONNX_LOGE("rule far: %s", f.c_str());
@@ -96,6 +99,34 @@ class OfflineTtsVitsImpl : public OfflineTtsImpl {
         tn_list_.push_back(std::make_unique<kaldifst::TextNormalizer>(is));
       }
     }
+
+    if (!config.rule_fars.empty()) {
+      std::vector<std::string> files;
+      SplitStringToVector(config.rule_fars, ",", false, &files);
+      tn_list_.reserve(files.size() + tn_list_.size());
+
+      for (const auto &f : files) {
+        if (config.model.debug) {
+          SHERPA_ONNX_LOGE("rule far: %s", f.c_str());
+        }
+
+        auto buf = ReadFile(mgr, f);
+
+        std::unique_ptr<std::istream> s(
+            new std::istrstream(buf.data(), buf.size()));
+
+        std::unique_ptr<fst::FarReader<fst::StdArc>> reader(
+            fst::FarReader<fst::StdArc>::Open(std::move(s)));
+
+        for (; !reader->Done(); reader->Next()) {
+          std::unique_ptr<fst::StdConstFst> r(
+              fst::CastOrConvertToConstFst(reader->GetFst()->Copy()));
+
+          tn_list_.push_back(
+              std::make_unique<kaldifst::TextNormalizer>(std::move(r)));
+        }  // for (; !reader->Done(); reader->Next())
+      }    // for (const auto &f : files)
+    }      // if (!config.rule_fars.empty())
   }
 #endif
 

sherpa-onnx/python/csrc/offline-stream.cc

@@ -4,6 +4,8 @@
 
 #include "sherpa-onnx/python/csrc/offline-stream.h"
 
+#include <vector>
+
 #include "sherpa-onnx/csrc/offline-stream.h"
 
 namespace sherpa_onnx {

sherpa-onnx/python/csrc/online-stream.cc

@@ -4,6 +4,8 @@
 
 #include "sherpa-onnx/python/csrc/online-stream.h"
 
+#include <vector>
+
 #include "sherpa-onnx/csrc/online-stream.h"
 
 namespace sherpa_onnx {