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

Support building GPU-capable sherpa-onnx on Linux aarch64. (#1500)

Browse Source 
Thanks to @Peakyxh for providing pre-built onnxruntime libraries 
with CUDA support for Linux aarch64.

Tested on Jetson nano b01
This commit is contained in:
Fangjun Kuang
2024-11-01 11:16:28 +08:00
committed by GitHub
parent a3c89aa0d8
commit 9ab89c33bc
41 changed files with 537 additions and 291 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
sherpa-onnx/csrc/online-zipformer2-ctc-model.cc
View File

@@ -70,7 +70,7 @@ class OnlineZipformer2CtcModel::Impl {
int32_t ChunkShift() const { return decode_chunk_len_; }
OrtAllocator *Allocator() const { return allocator_; }
OrtAllocator *Allocator() { return allocator_; }
// Return a vector containing 3 tensors
// - attn_cache
@@ -86,7 +86,7 @@ class OnlineZipformer2CtcModel::Impl {
}
std::vector<Ort::Value> StackStates(
std::vector<std::vector<Ort::Value>> states) const {
std::vector<std::vector<Ort::Value>> states) {
int32_t batch_size = static_cast<int32_t>(states.size());
std::vector<const Ort::Value *> buf(batch_size);
@@ -159,7 +159,7 @@ class OnlineZipformer2CtcModel::Impl {
}
std::vector<std::vector<Ort::Value>> UnStackStates(
std::vector<Ort::Value> states) const {
std::vector<Ort::Value> states) {
int32_t m = std::accumulate(num_encoder_layers_.begin(),
num_encoder_layers_.end(), 0);
assert(states.size() == m * 6 + 2);