EngineX-Iluvatar/enginex-mr_series-sherpa-onnx
Archived
16
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

Fix: Prepend 0 to tokenization to prevent word skipping for Kokoro. (#1787)

Browse Source 
Addressed issue Skipping words #1777
This commit is contained in:
ahadjawaid
2025-02-02 23:49:42 -06:00
committed by GitHub
parent 1d950a88e5
commit 8677d83efc
1 changed files with 2 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
sherpa-onnx/csrc/piper-phonemize-lexicon.cc
View File

@@ -163,6 +163,8 @@ static std::vector<std::vector<int64_t>> PiperPhonemesToIdsKokoro(
std::vector<int64_t> current;
current.reserve(phonemes.size());
current.push_back(0);
for (auto p : phonemes) {
if (token2id.count(p)) {
if (current.size() > max_len - 1) {