experiments/STT
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Make Alphabet copyable and default-constructable and avoid pointers

Browse Source
This commit is contained in:
Reuben Morais 2019-08-22 21:17:23 +02:00
parent 4d882a8aec
commit 4dabd248bc
8 changed files with 52 additions and 33 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
native_client/alphabet.h
View File

@ -15,7 +15,14 @@
*/
class Alphabet {
public:
Alphabet() {
}
Alphabet(const char *config_file) {
init(config_file);
}
void init(const char *config_file) {
std::ifstream in(config_file, std::ios::in);
unsigned int label = 0;
space_label_ = -2;

31
native_client/ctcdecode/scorer.cpp
View File

@ -30,18 +30,16 @@ static const int32_t MAGIC = 'TRIE';
static const int32_t FILE_VERSION = 4;
Scorer::Scorer(double alpha,
double beta,
const std::string& lm_path,
const std::string& trie_path,
const Alphabet& alphabet)
: dictionary()
, language_model_()
, is_character_based_(true)
double beta)
: is_character_based_(true)
, max_order_(0)
, alphabet_(alphabet)
{
reset_params(alpha, beta);
}
void
Scorer::init(const std::string& lm_path, const std::string& trie_path)
{
char_map_.clear();
SPACE_ID_ = alphabet_.GetSpaceLabel();
@ -60,9 +58,22 @@ Scorer::Scorer(double alpha,
double beta,
const std::string& lm_path,
const std::string& trie_path,
const std::string& alphabet_config_path)
: Scorer(alpha, beta, lm_path, trie_path, Alphabet(alphabet_config_path.c_str()))
const Alphabet& alphabet)
: Scorer(alpha, beta)
{
alphabet_ = alphabet;
init(lm_path, trie_path);
}
Scorer::Scorer(double alpha,
double beta,
const std::string& lm_path,
const std::string& trie_path,
const std::string& alphabet_config_path)
: Scorer(alpha, beta)
{
alphabet_.init(alphabet_config_path.c_str());
init(lm_path, trie_path);
}
Scorer::~Scorer()

5
native_client/ctcdecode/scorer.h
View File

@ -87,6 +87,11 @@ public:
std::unique_ptr<FstType> dictionary;
protected:
Scorer(double alpha,
double beta);
void init(const std::string& lm_path, const std::string& trie_path);
// necessary setup: load language model, fill FST's dictionary
void setup(const std::string &lm_path, const std::string &trie_path);

14
native_client/deepspeech.cc
View File

@ -244,7 +244,7 @@ StreamingState::processBatch(const vector<float>& buf, unsigned int n_steps)
previous_state_c_,
previous_state_h_);
const size_t num_classes = model_->alphabet_->GetSize() + 1; // +1 for blank
const size_t num_classes = model_->alphabet_.GetSize() + 1; // +1 for blank
const int n_frames = logits.size() / (ModelState::BATCH_SIZE * num_classes);
// Convert logits to double
@ -309,10 +309,10 @@ DS_EnableDecoderWithLM(ModelState* aCtx,
float aLMBeta)
{
try {
aCtx->scorer_ = new Scorer(aLMAlpha, aLMBeta,
aLMPath ? aLMPath : "",
aTriePath ? aTriePath : "",
*aCtx->alphabet_);
aCtx->scorer_.reset(new Scorer(aLMAlpha, aLMBeta,
aLMPath ? aLMPath : "",
aTriePath ? aTriePath : "",
aCtx->alphabet_));
return DS_ERR_OK;
} catch (...) {
return DS_ERR_INVALID_LM;
@ -343,11 +343,11 @@ DS_SetupStream(ModelState* aCtx,
const int cutoff_top_n = 40;
const double cutoff_prob = 1.0;
ctx->decoder_state_.init(*aCtx->alphabet_,
ctx->decoder_state_.init(aCtx->alphabet_,
aCtx->beam_width_,
cutoff_prob,
cutoff_top_n,
aCtx->scorer_);
aCtx->scorer_.get());
*retval = ctx.release();
return DS_ERR_OK;

12
native_client/modelstate.cc
View File

@ -7,9 +7,7 @@
using std::vector;
ModelState::ModelState()
: alphabet_(nullptr)
, scorer_(nullptr)
, beam_width_(-1)
: beam_width_(-1)
, n_steps_(-1)
, n_context_(-1)
, n_features_(-1)
@ -23,8 +21,6 @@ ModelState::ModelState()
ModelState::~ModelState()
{
delete scorer_;
delete alphabet_;
}
int
@ -36,7 +32,7 @@ ModelState::init(const char* model_path,
{
n_features_ = n_features;
n_context_ = n_context;
alphabet_ = new Alphabet(alphabet_path);
alphabet_.init(alphabet_path);
beam_width_ = beam_width;
return DS_ERR_OK;
}
@ -45,7 +41,7 @@ char*
ModelState::decode(const DecoderState& state)
{
vector<Output> out = state.decode();
return strdup(alphabet_->LabelsToString(out[0].tokens).c_str());
return strdup(alphabet_.LabelsToString(out[0].tokens).c_str());
}
Metadata*
@ -61,7 +57,7 @@ ModelState::decode_metadata(const DecoderState& state)
// Loop through each character
for (int i = 0; i < out[0].tokens.size(); ++i) {
items[i].character = strdup(alphabet_->StringFromLabel(out[0].tokens[i]).c_str());
items[i].character = strdup(alphabet_.StringFromLabel(out[0].tokens[i]).c_str());
items[i].timestep = out[0].timesteps[i];
items[i].start_time = out[0].timesteps[i] * ((float)audio_win_step_ / sample_rate_);

4
native_client/modelstate.h
View File

@ -19,8 +19,8 @@ struct ModelState {
static constexpr unsigned int DEFAULT_WINDOW_LENGTH = DEFAULT_SAMPLE_RATE * 0.032;
static constexpr unsigned int DEFAULT_WINDOW_STEP = DEFAULT_SAMPLE_RATE * 0.02;
Alphabet* alphabet_;
Scorer* scorer_;
Alphabet alphabet_;
std::unique_ptr<Scorer> scorer_;
unsigned int beam_width_;
unsigned int n_steps_;
unsigned int n_context_;

6
native_client/tflitemodelstate.cc
View File

@ -151,9 +151,9 @@ TFLiteModelState::init(const char* model_path,
TfLiteIntArray* dims_logits = interpreter_->tensor(logits_idx_)->dims;
const int final_dim_size = dims_logits->data[1] - 1;
if (final_dim_size != alphabet_->GetSize()) {
if (final_dim_size != alphabet_.GetSize()) {
std::cerr << "Error: Alphabet size does not match loaded model: alphabet "
<< "has size " << alphabet_->GetSize()
<< "has size " << alphabet_.GetSize()
<< ", but model has " << final_dim_size
<< " classes in its output. Make sure you're passing an alphabet "
<< "file with the same size as the one used for training."
@ -208,7 +208,7 @@ TFLiteModelState::infer(const vector<float>& mfcc,
vector<float>& state_c_output,
vector<float>& state_h_output)
{
const size_t num_classes = alphabet_->GetSize() + 1; // +1 for blank
const size_t num_classes = alphabet_.GetSize() + 1; // +1 for blank
// Feeding input_node
copy_vector_to_tensor(mfcc, input_node_idx_, n_frames*mfcc_feats_per_timestep_);

6
native_client/tfmodelstate.cc
View File

@ -108,9 +108,9 @@ TFModelState::init(const char* model_path,
}
int final_dim_size = logits_shape.vec<int>()(2) - 1;
if (final_dim_size != alphabet_->GetSize()) {
if (final_dim_size != alphabet_.GetSize()) {
std::cerr << "Error: Alphabet size does not match loaded model: alphabet "
<< "has size " << alphabet_->GetSize()
<< "has size " << alphabet_.GetSize()
<< ", but model has " << final_dim_size
<< " classes in its output. Make sure you're passing an alphabet "
<< "file with the same size as the one used for training."
@ -173,7 +173,7 @@ TFModelState::infer(const std::vector<float>& mfcc,
vector<float>& state_c_output,
vector<float>& state_h_output)
{
const size_t num_classes = alphabet_->GetSize() + 1; // +1 for blank
const size_t num_classes = alphabet_.GetSize() + 1; // +1 for blank
Tensor input = tensor_from_vector(mfcc, TensorShape({BATCH_SIZE, n_steps_, 2*n_context_+1, n_features_}));
Tensor previous_state_c_t = tensor_from_vector(previous_state_c, TensorShape({BATCH_SIZE, (long long)state_size_}));