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

Support VITS VCTK models (#367)

Browse Source 
* Support VITS VCTK models

* Release v1.8.1
This commit is contained in:
Fangjun Kuang
2023-10-16 17:22:30 +08:00
committed by GitHub
parent d01682d968
commit 9efe69720d
16 changed files with 332 additions and 31 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
sherpa-onnx/csrc/offline-tts-vits-impl.h
View File

@@ -23,7 +23,8 @@ class OfflineTtsVitsImpl : public OfflineTtsImpl {
lexicon_(config.model.vits.lexicon, config.model.vits.tokens,
model_->Punctuations()) {}
GeneratedAudio Generate(const std::string &text) const override {
GeneratedAudio Generate(const std::string &text,
int64_t sid = 0) const override {
std::vector<int64_t> x = lexicon_.ConvertTextToTokenIds(text);
if (x.empty()) {
SHERPA_ONNX_LOGE("Failed to convert %s to token IDs", text.c_str());
@@ -47,7 +48,7 @@ class OfflineTtsVitsImpl : public OfflineTtsImpl {
Ort::Value x_tensor = Ort::Value::CreateTensor(
memory_info, x.data(), x.size(), x_shape.data(), x_shape.size());
Ort::Value audio = model_->Run(std::move(x_tensor));
Ort::Value audio = model_->Run(std::move(x_tensor), sid);
std::vector<int64_t> audio_shape =
audio.GetTensorTypeAndShapeInfo().GetShape();