Update to Rust programming language.

src-tauri/vosk/model_small/README

@@ -0,0 +1,8 @@
+Small Russian model for Vosk (Android, RPi, other small devices)
+
+%WER 22.71 [ 9092 / 40042, 1124 ins, 1536 del, 6432 sub ] exp/chain_a/tdnn/decode_test_audiobooks_look_fast/wer_10_0.0
+%WER 11.79 [ 5940 / 50394, 894 ins, 832 del, 4214 sub ] exp/chain_a/tdnn/decode_test_golos_crowd_look_fast/wer_11_0.0
+%WER 21.34 [ 1789 / 8382, 173 ins, 440 del, 1176 sub ] exp/chain_a/tdnn/decode_test_golos_farfield_look_fast/wer_10_0.0
+%WER 29.89 [ 5579 / 18666, 476 ins, 1550 del, 3553 sub ] exp/chain_a/tdnn/decode_test_sova_devices_look_fast/wer_10_0.0
+%WER 31.97 [ 13588 / 42496, 1013 ins, 3640 del, 8935 sub ] exp/chain_a/tdnn/decode_test_youtube_look_fast/wer_9_0.0
+

src-tauri/vosk/model_small/conf/mfcc.conf

@@ -0,0 +1,7 @@
+--sample-frequency=16000
+--use-energy=false
+--num-mel-bins=40
+--num-ceps=40
+--low-freq=20
+--high-freq=7600
+--allow-downsample=true

src-tauri/vosk/model_small/conf/model.conf

@@ -0,0 +1,10 @@
+--min-active=200
+--max-active=3000
+--beam=10.0
+--lattice-beam=2.0
+--acoustic-scale=1.0
+--frame-subsampling-factor=3
+--endpoint.silence-phones=1:2:3:4:5:6:7:8:9:10
+--endpoint.rule2.min-trailing-silence=0.5
+--endpoint.rule3.min-trailing-silence=1.0
+--endpoint.rule4.min-trailing-silence=2.0

src-tauri/vosk/model_small/graph/disambig_tid.int

@@ -0,0 +1,5 @@
+9855
+9856
+9857
+9858
+9859

src-tauri/vosk/model_small/graph/phones/word_boundary.int

@@ -0,0 +1,202 @@
+1 nonword
+2 begin
+3 end
+4 internal
+5 singleton
+6 nonword
+7 begin
+8 end
+9 internal
+10 singleton
+11 begin
+12 end
+13 internal
+14 singleton
+15 begin
+16 end
+17 internal
+18 singleton
+19 begin
+20 end
+21 internal
+22 singleton
+23 begin
+24 end
+25 internal
+26 singleton
+27 begin
+28 end
+29 internal
+30 singleton
+31 begin
+32 end
+33 internal
+34 singleton
+35 begin
+36 end
+37 internal
+38 singleton
+39 begin
+40 end
+41 internal
+42 singleton
+43 begin
+44 end
+45 internal
+46 singleton
+47 begin
+48 end
+49 internal
+50 singleton
+51 begin
+52 end
+53 internal
+54 singleton
+55 begin
+56 end
+57 internal
+58 singleton
+59 begin
+60 end
+61 internal
+62 singleton
+63 begin
+64 end
+65 internal
+66 singleton
+67 begin
+68 end
+69 internal
+70 singleton
+71 begin
+72 end
+73 internal
+74 singleton
+75 begin
+76 end
+77 internal
+78 singleton
+79 begin
+80 end
+81 internal
+82 singleton
+83 begin
+84 end
+85 internal
+86 singleton
+87 begin
+88 end
+89 internal
+90 singleton
+91 begin
+92 end
+93 internal
+94 singleton
+95 begin
+96 end
+97 internal
+98 singleton
+99 begin
+100 end
+101 internal
+102 singleton
+103 begin
+104 end
+105 internal
+106 singleton
+107 begin
+108 end
+109 internal
+110 singleton
+111 begin
+112 end
+113 internal
+114 singleton
+115 begin
+116 end
+117 internal
+118 singleton
+119 begin
+120 end
+121 internal
+122 singleton
+123 begin
+124 end
+125 internal
+126 singleton
+127 begin
+128 end
+129 internal
+130 singleton
+131 begin
+132 end
+133 internal
+134 singleton
+135 begin
+136 end
+137 internal
+138 singleton
+139 begin
+140 end
+141 internal
+142 singleton
+143 begin
+144 end
+145 internal
+146 singleton
+147 begin
+148 end
+149 internal
+150 singleton
+151 begin
+152 end
+153 internal
+154 singleton
+155 begin
+156 end
+157 internal
+158 singleton
+159 begin
+160 end
+161 internal
+162 singleton
+163 begin
+164 end
+165 internal
+166 singleton
+167 begin
+168 end
+169 internal
+170 singleton
+171 begin
+172 end
+173 internal
+174 singleton
+175 begin
+176 end
+177 internal
+178 singleton
+179 begin
+180 end
+181 internal
+182 singleton
+183 begin
+184 end
+185 internal
+186 singleton
+187 begin
+188 end
+189 internal
+190 singleton
+191 begin
+192 end
+193 internal
+194 singleton
+195 begin
+196 end
+197 internal
+198 singleton
+199 begin
+200 end
+201 internal
+202 singleton

src-tauri/vosk/model_small/ivector/global_cmvn.stats

@@ -0,0 +1,3 @@
+ [
+  8.330133e+10 -4.600894e+09 -2.394861e+09 2.127165e+09 -9.355799e+09 -9.378007e+09 -1.302309e+10 -9.460417e+09 -9.260028e+09 -4.58608e+09 -5.287111e+09 -1.972033e+09 -6.090821e+09 -1.336419e+09 -5.214569e+09 -2.321841e+09 -3.889789e+09 -1.060202e+09 -2.065653e+09 -2.684904e+08 -7.4007e+08 -4587485 -1.315853e+08 -8597548 2.599227e+08 7.408538e+07 5.505751e+08 -1.161846e+07 5.138103e+08 -1.828159e+08 4.251498e+08 -2.901496e+07 6.469246e+08 2.489644e+08 6.289868e+08 2.490337e+08 3.38884e+08 -1.788837e+08 -2.536016e+08 -1.591728e+08 8.388078e+08 
+  8.660994e+12 4.637783e+11 3.366465e+11 4.467952e+11 5.094759e+11 5.179353e+11 6.145244e+11 4.970492e+11 5.014889e+11 4.027981e+11 3.937422e+11 3.602942e+11 3.162307e+11 2.40687e+11 2.267307e+11 1.563018e+11 1.341105e+11 8.535779e+10 6.12398e+10 3.207774e+10 1.737325e+10 5.704115e+09 7.980573e+08 2.168777e+08 2.763352e+09 6.859176e+09 1.214891e+10 1.604714e+10 2.005353e+10 2.240119e+10 2.366007e+10 2.300222e+10 2.406182e+10 2.354406e+10 2.098983e+10 1.619869e+10 1.491578e+10 1.224871e+10 9.502735e+09 6.517532e+09 0 ]

src-tauri/vosk/model_small/ivector/online_cmvn.conf

@@ -0,0 +1 @@
+# configuration file for apply-cmvn-online, used in the script ../local/run_online_decoding.sh

src-tauri/vosk/model_small/ivector/splice.conf

@@ -0,0 +1,2 @@
+--left-context=3
+--right-context=3

src-tauri/vosk/vosk_api.h

@@ -0,0 +1,361 @@
+// Copyright 2020-2021 Alpha Cephei Inc.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//       http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+/* This header contains the C API for Vosk speech recognition system */
+
+#ifndef VOSK_API_H
+#define VOSK_API_H
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/** Model stores all the data required for recognition
+ *  it contains static data and can be shared across processing
+ *  threads. */
+typedef struct VoskModel VoskModel;
+
+
+/** Speaker model is the same as model but contains the data
+ *  for speaker identification. */
+typedef struct VoskSpkModel VoskSpkModel;
+
+
+/** Recognizer object is the main object which processes data.
+ *  Each recognizer usually runs in own thread and takes audio as input.
+ *  Once audio is processed recognizer returns JSON object as a string
+ *  which represent decoded information - words, confidences, times, n-best lists,
+ *  speaker information and so on */
+typedef struct VoskRecognizer VoskRecognizer;
+
+
+/**
+ * Batch model object
+ */
+typedef struct VoskBatchModel VoskBatchModel;
+
+/**
+ * Batch recognizer object
+ */
+typedef struct VoskBatchRecognizer VoskBatchRecognizer;
+
+
+/** Loads model data from the file and returns the model object
+ *
+ * @param model_path: the path of the model on the filesystem
+ * @returns model object or NULL if problem occured */
+VoskModel *vosk_model_new(const char *model_path);
+
+
+/** Releases the model memory
+ *
+ *  The model object is reference-counted so if some recognizer
+ *  depends on this model, model might still stay alive. When
+ *  last recognizer is released, model will be released too. */
+void vosk_model_free(VoskModel *model);
+
+
+/** Check if a word can be recognized by the model
+ * @param word: the word
+ * @returns the word symbol if @param word exists inside the model
+ * or -1 otherwise.
+ * Reminding that word symbol 0 is for <epsilon> */
+int vosk_model_find_word(VoskModel *model, const char *word);
+
+
+/** Loads speaker model data from the file and returns the model object
+ *
+ * @param model_path: the path of the model on the filesystem
+ * @returns model object or NULL if problem occured */
+VoskSpkModel *vosk_spk_model_new(const char *model_path);
+
+
+/** Releases the model memory
+ *
+ *  The model object is reference-counted so if some recognizer
+ *  depends on this model, model might still stay alive. When
+ *  last recognizer is released, model will be released too. */
+void vosk_spk_model_free(VoskSpkModel *model);
+
+/** Creates the recognizer object
+ *
+ *  The recognizers process the speech and return text using shared model data
+ *  @param model       VoskModel containing static data for recognizer. Model can be
+ *                     shared across recognizers, even running in different threads.
+ *  @param sample_rate The sample rate of the audio you going to feed into the recognizer.
+ *                     Make sure this rate matches the audio content, it is a common
+ *                     issue causing accuracy problems.
+ *  @returns recognizer object or NULL if problem occured */
+VoskRecognizer *vosk_recognizer_new(VoskModel *model, float sample_rate);
+
+
+/** Creates the recognizer object with speaker recognition
+ *
+ *  With the speaker recognition mode the recognizer not just recognize
+ *  text but also return speaker vectors one can use for speaker identification
+ *
+ *  @param model       VoskModel containing static data for recognizer. Model can be
+ *                     shared across recognizers, even running in different threads.
+ *  @param sample_rate The sample rate of the audio you going to feed into the recognizer.
+ *                     Make sure this rate matches the audio content, it is a common
+ *                     issue causing accuracy problems.
+ *  @param spk_model speaker model for speaker identification
+ *  @returns recognizer object or NULL if problem occured */
+VoskRecognizer *vosk_recognizer_new_spk(VoskModel *model, float sample_rate, VoskSpkModel *spk_model);
+
+
+/** Creates the recognizer object with the phrase list
+ *
+ *  Sometimes when you want to improve recognition accuracy and when you don't need
+ *  to recognize large vocabulary you can specify a list of phrases to recognize. This
+ *  will improve recognizer speed and accuracy but might return [unk] if user said
+ *  something different.
+ *
+ *  Only recognizers with lookahead models support this type of quick configuration.
+ *  Precompiled HCLG graph models are not supported.
+ *
+ *  @param model       VoskModel containing static data for recognizer. Model can be
+ *                     shared across recognizers, even running in different threads.
+ *  @param sample_rate The sample rate of the audio you going to feed into the recognizer.
+ *                     Make sure this rate matches the audio content, it is a common
+ *                     issue causing accuracy problems.
+ *  @param grammar The string with the list of phrases to recognize as JSON array of strings,
+ *                 for example "["one two three four five", "[unk]"]".
+ *
+ *  @returns recognizer object or NULL if problem occured */
+VoskRecognizer *vosk_recognizer_new_grm(VoskModel *model, float sample_rate, const char *grammar);
+
+
+/** Adds speaker model to already initialized recognizer
+ *
+ * Can add speaker recognition model to already created recognizer. Helps to initialize
+ * speaker recognition for grammar-based recognizer.
+ *
+ * @param spk_model Speaker recognition model */
+void vosk_recognizer_set_spk_model(VoskRecognizer *recognizer, VoskSpkModel *spk_model);
+
+
+/** Reconfigures recognizer to use grammar
+ *
+ * @param recognizer   Already running VoskRecognizer
+ * @param grammar      Set of phrases in JSON array of strings or "[]" to use default model graph.
+ *                     See also vosk_recognizer_new_grm
+ */
+void vosk_recognizer_set_grm(VoskRecognizer *recognizer, char const *grammar);
+
+
+/** Configures recognizer to output n-best results
+ *
+ * <pre>
+ *   {
+ *      "alternatives": [
+ *          { "text": "one two three four five", "confidence": 0.97 },
+ *          { "text": "one two three for five", "confidence": 0.03 },
+ *      ]
+ *   }
+ * </pre>
+ *
+ * @param max_alternatives - maximum alternatives to return from recognition results
+ */
+void vosk_recognizer_set_max_alternatives(VoskRecognizer *recognizer, int max_alternatives);
+
+
+/** Enables words with times in the output
+ *
+ * <pre>
+ *   "result" : [{
+ *       "conf" : 1.000000,
+ *       "end" : 1.110000,
+ *       "start" : 0.870000,
+ *       "word" : "what"
+ *     }, {
+ *       "conf" : 1.000000,
+ *       "end" : 1.530000,
+ *       "start" : 1.110000,
+ *       "word" : "zero"
+ *     }, {
+ *       "conf" : 1.000000,
+ *       "end" : 1.950000,
+ *       "start" : 1.530000,
+ *       "word" : "zero"
+ *     }, {
+ *       "conf" : 1.000000,
+ *       "end" : 2.340000,
+ *       "start" : 1.950000,
+ *       "word" : "zero"
+ *     }, {
+ *       "conf" : 1.000000,
+ *       "end" : 2.610000,
+ *       "start" : 2.340000,
+ *       "word" : "one"
+ *     }],
+ * </pre>
+ *
+ * @param words - boolean value
+ */
+void vosk_recognizer_set_words(VoskRecognizer *recognizer, int words);
+
+/** Like above return words and confidences in partial results
+ *
+ * @param partial_words - boolean value
+ */
+void vosk_recognizer_set_partial_words(VoskRecognizer *recognizer, int partial_words);
+
+/** Set NLSML output
+ * @param nlsml - boolean value
+ */
+void vosk_recognizer_set_nlsml(VoskRecognizer *recognizer, int nlsml);
+
+
+/** Accept voice data
+ *
+ *  accept and process new chunk of voice data
+ *
+ *  @param data - audio data in PCM 16-bit mono format
+ *  @param length - length of the audio data
+ *  @returns 1 if silence is occured and you can retrieve a new utterance with result method 
+ *           0 if decoding continues
+ *           -1 if exception occured */
+int vosk_recognizer_accept_waveform(VoskRecognizer *recognizer, const char *data, int length);
+
+
+/** Same as above but the version with the short data for language bindings where you have
+ *  audio as array of shorts */
+int vosk_recognizer_accept_waveform_s(VoskRecognizer *recognizer, const short *data, int length);
+
+
+/** Same as above but the version with the float data for language bindings where you have
+ *  audio as array of floats */
+int vosk_recognizer_accept_waveform_f(VoskRecognizer *recognizer, const float *data, int length);
+
+
+/** Returns speech recognition result
+ *
+ * @returns the result in JSON format which contains decoded line, decoded
+ *          words, times in seconds and confidences. You can parse this result
+ *          with any json parser
+ *
+ * <pre>
+ *  {
+ *    "text" : "what zero zero zero one"
+ *  }
+ * </pre>
+ *
+ * If alternatives enabled it returns result with alternatives, see also vosk_recognizer_set_max_alternatives().
+ *
+ * If word times enabled returns word time, see also vosk_recognizer_set_word_times().
+ */
+const char *vosk_recognizer_result(VoskRecognizer *recognizer);
+
+
+/** Returns partial speech recognition
+ *
+ * @returns partial speech recognition text which is not yet finalized.
+ *          result may change as recognizer process more data.
+ *
+ * <pre>
+ * {
+ *    "partial" : "cyril one eight zero"
+ * }
+ * </pre>
+ */
+const char *vosk_recognizer_partial_result(VoskRecognizer *recognizer);
+
+
+/** Returns speech recognition result. Same as result, but doesn't wait for silence
+ *  You usually call it in the end of the stream to get final bits of audio. It
+ *  flushes the feature pipeline, so all remaining audio chunks got processed.
+ *
+ *  @returns speech result in JSON format.
+ */
+const char *vosk_recognizer_final_result(VoskRecognizer *recognizer);
+
+
+/** Resets the recognizer
+ *
+ *  Resets current results so the recognition can continue from scratch */
+void vosk_recognizer_reset(VoskRecognizer *recognizer);
+
+
+/** Releases recognizer object
+ *
+ *  Underlying model is also unreferenced and if needed released */
+void vosk_recognizer_free(VoskRecognizer *recognizer);
+
+/** Set log level for Kaldi messages
+ *
+ *  @param log_level the level
+ *     0 - default value to print info and error messages but no debug
+ *     less than 0 - don't print info messages
+ *     greather than 0 - more verbose mode
+ */
+void vosk_set_log_level(int log_level);
+
+/**
+ *  Init, automatically select a CUDA device and allow multithreading.
+ *  Must be called once from the main thread.
+ *  Has no effect if HAVE_CUDA flag is not set.
+ */
+void vosk_gpu_init();
+
+/**
+ *  Init CUDA device in a multi-threaded environment.
+ *  Must be called for each thread.
+ *  Has no effect if HAVE_CUDA flag is not set.
+ */
+void vosk_gpu_thread_init();
+
+/** Creates the batch recognizer object
+ *
+ *  @returns model object or NULL if problem occured */
+VoskBatchModel *vosk_batch_model_new(const char *model_path);
+
+/** Releases batch model object */
+void vosk_batch_model_free(VoskBatchModel *model);
+
+/** Wait for the processing */
+void vosk_batch_model_wait(VoskBatchModel *model);
+
+/** Creates batch recognizer object
+ *  @returns recognizer object or NULL if problem occured */
+VoskBatchRecognizer *vosk_batch_recognizer_new(VoskBatchModel *model, float sample_rate);
+ 
+/** Releases batch recognizer object */
+void vosk_batch_recognizer_free(VoskBatchRecognizer *recognizer);
+
+/** Accept batch voice data */
+void vosk_batch_recognizer_accept_waveform(VoskBatchRecognizer *recognizer, const char *data, int length);
+
+/** Set NLSML output
+ * @param nlsml - boolean value
+ */
+void vosk_batch_recognizer_set_nlsml(VoskBatchRecognizer *recognizer, int nlsml);
+
+/** Closes the stream */
+void vosk_batch_recognizer_finish_stream(VoskBatchRecognizer *recognizer);
+
+/** Return results */
+const char *vosk_batch_recognizer_front_result(VoskBatchRecognizer *recognizer);
+
+/** Release and free first retrieved result */
+void vosk_batch_recognizer_pop(VoskBatchRecognizer *recognizer);
+
+/** Get amount of pending chunks for more intelligent waiting */
+int vosk_batch_recognizer_get_pending_chunks(VoskBatchRecognizer *recognizer);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* VOSK_API_H */