janitor

janitor
git clone https://git.prime8.dev/janitor.git
Log | Files | Refs | README | LICENSE

feature-fbank.h (4608B)

      1 /**
      2  * Copyright (c)  2022  Xiaomi Corporation (authors: Fangjun Kuang)
      3  *
      4  * See LICENSE for clarification regarding multiple authors
      5  *
      6  * Licensed under the Apache License, Version 2.0 (the "License");
      7  * you may not use this file except in compliance with the License.
      8  * You may obtain a copy of the License at
      9  *
     10  *     http://www.apache.org/licenses/LICENSE-2.0
     11  *
     12  * Unless required by applicable law or agreed to in writing, software
     13  * distributed under the License is distributed on an "AS IS" BASIS,
     14  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
     15  * See the License for the specific language governing permissions and
     16  * limitations under the License.
     17  */
     18 
     19 // This file is copied/modified from kaldi/src/feat/feature-fbank.h
     20 
     21 #ifndef KALDI_NATIVE_FBANK_CSRC_FEATURE_FBANK_H_
     22 #define KALDI_NATIVE_FBANK_CSRC_FEATURE_FBANK_H_
     23 
     24 #include <map>
     25 #include <string>
     26 #include <vector>
     27 #include <cstdint>
     28 #include <sstream>
     29 
     30 #include "kaldi-native-fbank/csrc/feature-window.h"
     31 #include "kaldi-native-fbank/csrc/mel-computations.h"
     32 #include "kaldi-native-fbank/csrc/rfft.h"
     33 
     34 namespace knf {
     35 
     36 struct FbankOptions {
     37   FrameExtractionOptions frame_opts;
     38   MelBanksOptions mel_opts;
     39   // append an extra dimension with energy to the filter banks
     40   bool use_energy = false;
     41   float energy_floor = 0.0f;  // active iff use_energy==true
     42 
     43   // If true, compute log_energy before preemphasis and windowing
     44   // If false, compute log_energy after preemphasis ans windowing
     45   bool raw_energy = true;  // active iff use_energy==true
     46 
     47   // If true, put energy last (if using energy)
     48   // If false, put energy first
     49   bool htk_compat = false;  // active iff use_energy==true
     50 
     51   // if true (default), produce log-filterbank, else linear
     52   bool use_log_fbank = true;
     53 
     54   // if true (default), use power in filterbank
     55   // analysis, else magnitude.
     56   bool use_power = true;
     57 
     58   FbankOptions() { mel_opts.num_bins = 23; }
     59 
     60   std::string ToString() const {
     61     std::ostringstream os;
     62     os << "frame_opts: \n";
     63     os << frame_opts << "\n";
     64     os << "\n";
     65 
     66     os << "mel_opts: \n";
     67     os << mel_opts << "\n";
     68 
     69     os << "use_energy: " << use_energy << "\n";
     70     os << "energy_floor: " << energy_floor << "\n";
     71     os << "raw_energy: " << raw_energy << "\n";
     72     os << "htk_compat: " << htk_compat << "\n";
     73     os << "use_log_fbank: " << use_log_fbank << "\n";
     74     os << "use_power: " << use_power << "\n";
     75     return os.str();
     76   }
     77 };
     78 
     79 std::ostream &operator<<(std::ostream &os, const FbankOptions &opts);
     80 
     81 class FbankComputer {
     82  public:
     83   using Options = FbankOptions;
     84 
     85   explicit FbankComputer(const FbankOptions &opts);
     86   ~FbankComputer();
     87 
     88   int32_t Dim() const {
     89     return opts_.mel_opts.num_bins + (opts_.use_energy ? 1 : 0);
     90   }
     91 
     92   // if true, compute log_energy_pre_window but after dithering and dc removal
     93   bool NeedRawLogEnergy() const { return opts_.use_energy && opts_.raw_energy; }
     94 
     95   const FrameExtractionOptions &GetFrameOptions() const {
     96     return opts_.frame_opts;
     97   }
     98 
     99   const FbankOptions &GetOptions() const { return opts_; }
    100 
    101   /**
    102      Function that computes one frame of features from
    103      one frame of signal.
    104 
    105      @param [in] signal_raw_log_energy The log-energy of the frame of the signal
    106          prior to windowing and pre-emphasis, or
    107          log(numeric_limits<float>::min()), whichever is greater.  Must be
    108          ignored by this function if this class returns false from
    109          this->NeedsRawLogEnergy().
    110      @param [in] vtln_warp  The VTLN warping factor that the user wants
    111          to be applied when computing features for this utterance.  Will
    112          normally be 1.0, meaning no warping is to be done.  The value will
    113          be ignored for feature types that don't support VLTN, such as
    114          spectrogram features.
    115      @param [in] signal_frame  One frame of the signal,
    116        as extracted using the function ExtractWindow() using the options
    117        returned by this->GetFrameOptions().  The function will use the
    118        vector as a workspace, which is why it's a non-const pointer.
    119      @param [out] feature  Pointer to a vector of size this->Dim(), to which
    120          the computed feature will be written. It should be pre-allocated.
    121   */
    122   void Compute(float signal_raw_log_energy, float vtln_warp,
    123                std::vector<float> *signal_frame, float *feature);
    124 
    125  private:
    126   const MelBanks *GetMelBanks(float vtln_warp);
    127 
    128   FbankOptions opts_;
    129   float log_energy_floor_;
    130   std::map<float, MelBanks *> mel_banks_;  // float is VTLN coefficient.
    131   Rfft rfft_;
    132 };
    133 
    134 }  // namespace knf
    135 
    136 #endif  // KALDI_NATIVE_FBANK_CSRC_FEATURE_FBANK_H_