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

Add non-streaming ASR support for HarmonyOS. (#1564)

Browse Source
This commit is contained in:
Fangjun Kuang
2024-11-26 16:38:35 +08:00
committed by GitHub
parent a4b79f077c
commit 298b6b6fda
50 changed files with 648 additions and 454 deletions
Download Patch File Download Diff File Expand all files Collapse all files

47
sherpa-onnx/csrc/offline-whisper-model.cc
View File

@@ -11,6 +11,15 @@
#include <unordered_map>
#include <utility>
#if __ANDROID_API__ >= 9
#include "android/asset_manager.h"
#include "android/asset_manager_jni.h"
#endif
#if __OHOS__
#include "rawfile/raw_file_manager.h"
#endif
#include "sherpa-onnx/csrc/macros.h"
#include "sherpa-onnx/csrc/onnx-utils.h"
#include "sherpa-onnx/csrc/session.h"
@@ -52,8 +61,8 @@ class OfflineWhisperModel::Impl {
}
}
#if __ANDROID_API__ >= 9
Impl(AAssetManager *mgr, const OfflineModelConfig &config)
template <typename Manager>
Impl(Manager *mgr, const OfflineModelConfig &config)
: config_(config),
env_(ORT_LOGGING_LEVEL_ERROR),
sess_opts_(GetSessionOptions(config)),
@@ -69,7 +78,8 @@ class OfflineWhisperModel::Impl {
}
}
Impl(AAssetManager *mgr, const SpokenLanguageIdentificationConfig &config)
template <typename Manager>
Impl(Manager *mgr, const SpokenLanguageIdentificationConfig &config)
: lid_config_(config),
env_(ORT_LOGGING_LEVEL_ERROR),
sess_opts_(GetSessionOptions(config)),
@@ -84,7 +94,6 @@ class OfflineWhisperModel::Impl {
InitDecoder(buf.data(), buf.size());
}
}
#endif
std::pair<Ort::Value, Ort::Value> ForwardEncoder(Ort::Value features) {
auto encoder_out = encoder_sess_->Run(
@@ -237,7 +246,11 @@ class OfflineWhisperModel::Impl {
std::ostringstream os;
os << "---encoder---\n";
PrintModelMetadata(os, meta_data);
#if __OHOS__
SHERPA_ONNX_LOGE("%{public}s\n", os.str().c_str());
#else
SHERPA_ONNX_LOGE("%s\n", os.str().c_str());
#endif
}
Ort::AllocatorWithDefaultOptions allocator; // used in the macro below
@@ -338,17 +351,16 @@ OfflineWhisperModel::OfflineWhisperModel(
const SpokenLanguageIdentificationConfig &config)
: impl_(std::make_unique<Impl>(config)) {}
#if __ANDROID_API__ >= 9
OfflineWhisperModel::OfflineWhisperModel(AAssetManager *mgr,
template <typename Manager>
OfflineWhisperModel::OfflineWhisperModel(Manager *mgr,
const OfflineModelConfig &config)
: impl_(std::make_unique<Impl>(mgr, config)) {}
template <typename Manager>
OfflineWhisperModel::OfflineWhisperModel(
AAssetManager *mgr, const SpokenLanguageIdentificationConfig &config)
Manager *mgr, const SpokenLanguageIdentificationConfig &config)
: impl_(std::make_unique<Impl>(mgr, config)) {}
#endif
OfflineWhisperModel::~OfflineWhisperModel() = default;
std::pair<Ort::Value, Ort::Value> OfflineWhisperModel::ForwardEncoder(
@@ -453,4 +465,21 @@ void OfflineWhisperModel::NormalizeFeatures(float *features, int32_t num_frames,
}
}
#if __ANDROID_API__ >= 9
template OfflineWhisperModel::OfflineWhisperModel(
AAssetManager *mgr, const OfflineModelConfig &config);
template OfflineWhisperModel::OfflineWhisperModel(
AAssetManager *mgr, const SpokenLanguageIdentificationConfig &config);
#endif
#if __OHOS__
template OfflineWhisperModel::OfflineWhisperModel(
NativeResourceManager *mgr, const OfflineModelConfig &config);
template OfflineWhisperModel::OfflineWhisperModel(
NativeResourceManager *mgr,
const SpokenLanguageIdentificationConfig &config);
#endif
} // namespace sherpa_onnx