decoder for open vocabulary keyword spotting (#505)

* various fixes to ContextGraph to support open vocabulary keywords decoder

* Add keyword spotter runtime

* Add binary

* First version works

* Minor fixes

* update text2token

* default values

* Add jni for kws

* add kws android project

* Minor fixes

* Remove unused interface

* Minor fixes

* Add workflow

* handle extra info in texts

* Minor fixes

* Add more comments

* Fix ci

* fix cpp style

* Add input box in android demo so that users can specify their keywords

* Fix cpp style

* Fix comments

* Minor fixes

* Minor fixes

* minor fixes

* Minor fixes

* Minor fixes

* Add CI

* Fix code style

* cpplint

* Fix comments

* Fix error
This commit is contained in:
Wei Kang
2024-01-20 22:52:41 +08:00
committed by GitHub
parent bf1dd3daf6
commit b6c020901a
77 changed files with 3316 additions and 68 deletions

View File

@@ -14,9 +14,11 @@
#include "sherpa-onnx/csrc/online-ctc-decoder.h"
#include "sherpa-onnx/csrc/online-paraformer-decoder.h"
#include "sherpa-onnx/csrc/online-transducer-decoder.h"
#include "sherpa-onnx/csrc/transducer-keyword-decoder.h"
namespace sherpa_onnx {
class TransducerKeywordResult;
class OnlineStream {
public:
explicit OnlineStream(const FeatureExtractorConfig &config = {},
@@ -76,6 +78,9 @@ class OnlineStream {
void SetResult(const OnlineTransducerDecoderResult &r);
OnlineTransducerDecoderResult &GetResult();
void SetKeywordResult(const TransducerKeywordResult &r);
TransducerKeywordResult &GetKeywordResult(bool remove_duplicates = false);
void SetCtcResult(const OnlineCtcDecoderResult &r);
OnlineCtcDecoderResult &GetCtcResult();
@@ -92,7 +97,7 @@ class OnlineStream {
*/
const ContextGraphPtr &GetContextGraph() const;
// for streaming parformer
// for streaming paraformer
std::vector<float> &GetParaformerFeatCache();
std::vector<float> &GetParaformerEncoderOutCache();
std::vector<float> &GetParaformerAlphaCache();