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

Add C++ runtime for speaker verification models from NeMo (#527)

Browse Source
This commit is contained in:
Fangjun Kuang
2024-01-13 21:42:09 +08:00
committed by GitHub
parent 68a525a024
commit 2024e96639
20 changed files with 405 additions and 24 deletions
Download Patch File Download Diff File Expand all files Collapse all files

28
sherpa-onnx/csrc/speaker-embedding-extractor-nemo-model-meta-data.h Normal file
View File

@@ -0,0 +1,28 @@
// sherpa-onnx/csrc/speaker-embedding-extractor-nemo-model-meta-data.h
//
// Copyright (c) 2024 Xiaomi Corporation
#ifndef SHERPA_ONNX_CSRC_SPEAKER_EMBEDDING_EXTRACTOR_NEMO_MODEL_META_DATA_H_
#define SHERPA_ONNX_CSRC_SPEAKER_EMBEDDING_EXTRACTOR_NEMO_MODEL_META_DATA_H_
#include <cstdint>
#include <string>
namespace sherpa_onnx {
struct SpeakerEmbeddingExtractorNeMoModelMetaData {
int32_t output_dim = 0;
int32_t feat_dim = 80;
int32_t sample_rate = 0;
int32_t window_size_ms = 25;
int32_t window_stride_ms = 25;
// Chinese, English, etc.
std::string language;
// for 3d-speaker, it is global-mean
std::string feature_normalize_type;
std::string window_type = "hann";
};
} // namespace sherpa_onnx
#endif // SHERPA_ONNX_CSRC_SPEAKER_EMBEDDING_EXTRACTOR_NEMO_MODEL_META_DATA_H_