Add more debug info, untested

This commit is contained in:
msqr1
2024-03-16 23:23:05 -07:00
parent 9c261cb5eb
commit ecab9aa5d5
9 changed files with 5899 additions and 48 deletions

View File

@@ -1,18 +1,21 @@
# Overview # Overview
- **STILL UNDER DEVELOPMENT!**
- A speech recognizer built on Vosk that can be run on the browser, inspired by [vosk-browser](https://github.com/ccoreilly/vosk-browser), but built from scratch and no code taken! - A speech recognizer built on Vosk that can be run on the browser, inspired by [vosk-browser](https://github.com/ccoreilly/vosk-browser), but built from scratch and no code taken!
- Designed with basic/nothrow exception safety - Designed with basic/nothrow exception safety
- See the *usage* folder's README.md for API documentation, notes. - See the *usage* folder's README.md for API documentation, notes.
- See the *devel* folder for the newest build (not guaranteed to work) and the JS build script - See the *devel* folder for the newest build (not guaranteed to work) and the JS build script
# Additions to vosk-browser: # Additions to vosk-browser:
- Download multiple models - Support multiple models
- Model storage path management (for multiple models) - Model storage path management
- Model ID management (for model updates) - Model ID management (for model updates)
- Smaller JS size (>3.1MB vs 1.4MB) - Smaller JS size (>3.1MB vs 1.4MB)
- All related files (pthread worker, worklet processor,...) are merged - All related files (pthread worker, worklet processor,...) are merged
- Shorter from-scratch build time - Shorter from-scratch build time
# Basic usage # Disclaimer:
- Use the prebuilt binary through JsDelivr. - Will log empty statement into JS console, see *global.cc* line 34
- Model loading can be slow due due to slow OPFS reading speed
# Basic usage
- Use the prebuilt binary through JsDelivr

File diff suppressed because one or more lines are too long

View File

@@ -1 +1,165 @@
"use strict";var Module={};var initializedJS=false;function threadPrintErr(...args){var text=args.join(" ");console.error(text)}function threadAlert(...args){var text=args.join(" ");postMessage({cmd:"alert",text:text,threadId:Module["_pthread_self"]()})}var err=threadPrintErr;self.alert=threadAlert;Module["instantiateWasm"]=(info,receiveInstance)=>{var module=Module["wasmModule"];Module["wasmModule"]=null;var instance=new WebAssembly.Instance(module,info);return receiveInstance(instance)};self.onunhandledrejection=e=>{throw e.reason||e};function handleMessage(e){try{if(e.data.cmd==="load"){let messageQueue=[];self.onmessage=e=>messageQueue.push(e);self.startWorker=instance=>{Module=instance;postMessage({"cmd":"loaded"});for(let msg of messageQueue){handleMessage(msg)}self.onmessage=handleMessage};Module["wasmModule"]=e.data.wasmModule;for(const handler of e.data.handlers){Module[handler]=(...args)=>{postMessage({cmd:"callHandler",handler:handler,args:args})}}Module["wasmMemory"]=e.data.wasmMemory;Module["buffer"]=Module["wasmMemory"].buffer;Module["ENVIRONMENT_IS_PTHREAD"]=true;if(typeof e.data.urlOrBlob=="string"){importScripts(e.data.urlOrBlob)}else{var objectUrl=URL.createObjectURL(e.data.urlOrBlob);importScripts(objectUrl);URL.revokeObjectURL(objectUrl)}loadVosklet(Module)}else if(e.data.cmd==="run"){Module["__emscripten_thread_init"](e.data.pthread_ptr,0,0,1);Module["__emscripten_thread_mailbox_await"](e.data.pthread_ptr);Module["establishStackSpace"]();Module["PThread"].receiveObjectTransfer(e.data);Module["PThread"].threadInitTLS();if(!initializedJS){Module["__embind_initialize_bindings"]();initializedJS=true}try{Module["invokeEntryPoint"](e.data.start_routine,e.data.arg)}catch(ex){if(ex!="unwind"){throw ex}}}else if(e.data.cmd==="cancel"){if(Module["_pthread_self"]()){Module["__emscripten_thread_exit"](-1)}}else if(e.data.target==="setimmediate"){}else if(e.data.cmd==="checkMailbox"){if(initializedJS){Module["checkMailbox"]()}}else if(e.data.cmd){err(`worker.js received unknown command ${e.data.cmd}`);err(e.data)}}catch(ex){Module["__emscripten_thread_crashed"]?.();throw ex}}self.onmessage=handleMessage; /**
* @license
* Copyright 2015 The Emscripten Authors
* SPDX-License-Identifier: MIT
*/
// Pthread Web Worker startup routine:
// This is the entry point file that is loaded first by each Web Worker
// that executes pthreads on the Emscripten application.
'use strict';
var Module = {};
// Thread-local guard variable for one-time init of the JS state
var initializedJS = false;
function assert(condition, text) {
if (!condition) abort('Assertion failed: ' + text);
}
function threadPrintErr(...args) {
var text = args.join(' ');
console.error(text);
}
function threadAlert(...args) {
var text = args.join(' ');
postMessage({cmd: 'alert', text, threadId: Module['_pthread_self']()});
}
// We don't need out() for now, but may need to add it if we want to use it
// here. Or, if this code all moves into the main JS, that problem will go
// away. (For now, adding it here increases code size for no benefit.)
var out = () => { throw 'out() is not defined in worker.js.'; }
var err = threadPrintErr;
self.alert = threadAlert;
var dbg = threadPrintErr;
Module['instantiateWasm'] = (info, receiveInstance) => {
// Instantiate from the module posted from the main thread.
// We can just use sync instantiation in the worker.
var module = Module['wasmModule'];
// We don't need the module anymore; new threads will be spawned from the main thread.
Module['wasmModule'] = null;
var instance = new WebAssembly.Instance(module, info);
// TODO: Due to Closure regression https://github.com/google/closure-compiler/issues/3193,
// the above line no longer optimizes out down to the following line.
// When the regression is fixed, we can remove this if/else.
return receiveInstance(instance);
}
// Turn unhandled rejected promises into errors so that the main thread will be
// notified about them.
self.onunhandledrejection = (e) => {
throw e.reason || e;
};
function handleMessage(e) {
try {
if (e.data.cmd === 'load') { // Preload command that is called once per worker to parse and load the Emscripten code.
// Until we initialize the runtime, queue up any further incoming messages.
let messageQueue = [];
self.onmessage = (e) => messageQueue.push(e);
// And add a callback for when the runtime is initialized.
self.startWorker = (instance) => {
Module = instance;
// Notify the main thread that this thread has loaded.
postMessage({ 'cmd': 'loaded' });
// Process any messages that were queued before the thread was ready.
for (let msg of messageQueue) {
handleMessage(msg);
}
// Restore the real message handler.
self.onmessage = handleMessage;
};
// Module and memory were sent from main thread
Module['wasmModule'] = e.data.wasmModule;
// Use `const` here to ensure that the variable is scoped only to
// that iteration, allowing safe reference from a closure.
for (const handler of e.data.handlers) {
Module[handler] = (...args) => {
dbg(`calling handler on main thread: ${handler}`);
postMessage({ cmd: 'callHandler', handler, args: args });
}
}
Module['wasmMemory'] = e.data.wasmMemory;
Module['wasmOffsetData'] = e.data.wasmOffsetConverter;
Module['buffer'] = Module['wasmMemory'].buffer;
Module['workerID'] = e.data.workerID;
Module['ENVIRONMENT_IS_PTHREAD'] = true;
if (typeof e.data.urlOrBlob == 'string') {
importScripts(e.data.urlOrBlob);
} else {
var objectUrl = URL.createObjectURL(e.data.urlOrBlob);
importScripts(objectUrl);
URL.revokeObjectURL(objectUrl);
}
loadVosklet(Module);
} else if (e.data.cmd === 'run') {
// Pass the thread address to wasm to store it for fast access.
Module['__emscripten_thread_init'](e.data.pthread_ptr, /*is_main=*/0, /*is_runtime=*/0, /*can_block=*/1);
// Await mailbox notifications with `Atomics.waitAsync` so we can start
// using the fast `Atomics.notify` notification path.
Module['__emscripten_thread_mailbox_await'](e.data.pthread_ptr);
assert(e.data.pthread_ptr);
// Also call inside JS module to set up the stack frame for this pthread in JS module scope
Module['establishStackSpace']();
Module['PThread'].receiveObjectTransfer(e.data);
Module['PThread'].threadInitTLS();
if (!initializedJS) {
// Embind must initialize itself on all threads, as it generates support JS.
// We only do this once per worker since they get reused
Module['__embind_initialize_bindings']();
initializedJS = true;
}
try {
Module['invokeEntryPoint'](e.data.start_routine, e.data.arg);
} catch(ex) {
if (ex != 'unwind') {
// The pthread "crashed". Do not call `_emscripten_thread_exit` (which
// would make this thread joinable). Instead, re-throw the exception
// and let the top level handler propagate it back to the main thread.
throw ex;
}
dbg(`Pthread 0x${Module['_pthread_self']().toString(16)} completed its main entry point with an 'unwind', keeping the worker alive for asynchronous operation.`);
}
} else if (e.data.cmd === 'cancel') { // Main thread is asking for a pthread_cancel() on this thread.
if (Module['_pthread_self']()) {
Module['__emscripten_thread_exit'](-1);
}
} else if (e.data.target === 'setimmediate') {
// no-op
} else if (e.data.cmd === 'checkMailbox') {
if (initializedJS) {
Module['checkMailbox']();
}
} else if (e.data.cmd) {
// The received message looks like something that should be handled by this message
// handler, (since there is a e.data.cmd field present), but is not one of the
// recognized commands:
err(`worker.js received unknown command ${e.data.cmd}`);
err(e.data);
}
} catch(ex) {
err(`worker.js onmessage() captured an uncaught exception: ${ex}`);
if (ex?.stack) err(ex.stack);
Module['__emscripten_thread_crashed']?.();
throw ex;
}
};
self.onmessage = handleMessage;

View File

@@ -4,9 +4,27 @@
<script src="Vosklet.js"></script> <script src="Vosklet.js"></script>
<script> <script>
window.onload = async () => { window.onload = async () => {
const module = await loadVosklet(); window.ctx = new AudioContext()
const model = await module.makeModel("../usage/en-model.tgz","model","ID"); let micNode = ctx.createMediaStreamSource(await navigator.mediaDevices.getUserMedia({
const recognizer = await module.makeRecognizer() video: false,
audio: {
echoCancellation: true,
noiseSuppression: true,
channelCount: 1,
sampleRate: 16000
},
}))
window.module = await loadVosklet()
window.model = await module.makeModel("../usage/en-model.tgz","model","ID")
window.recognizer = await module.makeRecognizer(model, ctx.sampleRate)
recognizer.addEventListener("result", e => {
console.log("Result: ", e.detail)
})
recognizer.addEventListener("partialResult", e => {
console.log("Partial result: ", e.detail)
})
//let recNode = recognizer.getNode(ctx);
} }
</script> </script>
</head> </head>

View File

@@ -8,7 +8,7 @@ OPENFST=$(realpath openfst) &&
LIBARCHIVE=$(realpath libarchive) && LIBARCHIVE=$(realpath libarchive) &&
CLAPACK_WASM=$(realpath clapack-wasm) && CLAPACK_WASM=$(realpath clapack-wasm) &&
MAX_MEMORY=${MAX_MEMORY:-300mb} && MAX_MEMORY=${MAX_MEMORY:-500mb} &&
MAX_THREADS=${MAX_THREADS:-2} && MAX_THREADS=${MAX_THREADS:-2} &&
EMSDK=${EMSDK:-$(realpath emsdk)} && EMSDK=${EMSDK:-$(realpath emsdk)} &&
@@ -28,12 +28,12 @@ fi
. $EMSDK/emsdk_env.sh && . $EMSDK/emsdk_env.sh &&
cd $SRC && cd $SRC &&
MODE=1 && # 0: Ultra debug info, 1: Extremely optimized release, else custom MODE=2 && # 0: Ultra debug info, 1: Extremely optimized release, else custom
echo "Mode = $MODE" && echo "Mode = $MODE" &&
if [ $MODE = 0 ]; then if [ $MODE = 0 ]; then
em++ -O0 global.cc genericModel.cc recognizer.cc bindings.cc -sWASMFS -sWASM_BIGINT -sSINGLE_FILE -sGZIP_EMBEDDINGS -sMODULARIZE -sEMBIND_STD_STRING_IS_UTF8 -sPTHREAD_POOL_DELAY_LOAD -sRUNTIME_DEBUG -sSTACK_OVERFLOW_CHECK=2 -sTEXTDECODER=2 -sPTHREAD_POOL_SIZE_STRICT=2 -sASSERTIONS=2 -sINITIAL_MEMORY=$MAX_MEMORY -sPTHREAD_POOL_SIZE=$MAX_THREADS -sDISABLE_EXCEPTION_CATCHING=0 -sEXIT_RUNTIME=0 -sINVOKE_RUN=0 -sPOLYFILL=0 -sEXPORTED_FUNCTIONS=_malloc -sEXPORT_NAME=loadVosklet -sMALLOC=emmalloc -sEXPORTED_RUNTIME_METHODS=UTF8ToString,stringToUTF8OnStack -sENVIRONMENT=web,worker -I. -I$LIBARCHIVE/include -I$VOSK/src -L$LIBARCHIVE/lib -larchive -L$KALDI/src -l:online2/kaldi-online2.a -l:decoder/kaldi-decoder.a -l:ivector/kaldi-ivector.a -l:gmm/kaldi-gmm.a -l:tree/kaldi-tree.a -l:feat/kaldi-feat.a -l:cudamatrix/kaldi-cudamatrix.a -l:lat/kaldi-lat.a -l:lm/kaldi-lm.a -l:rnnlm/kaldi-rnnlm.a -l:hmm/kaldi-hmm.a -l:nnet3/kaldi-nnet3.a -l:transform/kaldi-transform.a -l:matrix/kaldi-matrix.a -l:fstext/kaldi-fstext.a -l:util/kaldi-util.a -l:base/kaldi-base.a -L$OPENFST/lib -l:libfst.a -l:libfstngram.a -L$CLAPACK_WASM -l:CBLAS/lib/cblas.a -l:CLAPACK-3.2.1/lapack.a -l:CLAPACK-3.2.1/libcblaswr.a -l:f2c_BLAS-3.8.0/blas.a -l:libf2c/libf2c.a -L$VOSK/src -l:vosk.a -lopfs.js -lembind -pthread -flto -msimd128 --embind-emit-tsd Vosklet.d.ts -fsanitize=undefined -fsanitize=address -fsanitize=leak -g3 --pre-js pre.js -o ../devel/Vosklet.js em++ -O0 global.cc genericModel.cc recognizer.cc bindings.cc -sWASMFS -sWASM_BIGINT -sSINGLE_FILE -sGZIP_EMBEDDINGS -sMODULARIZE -sEMBIND_STD_STRING_IS_UTF8 -sPTHREAD_POOL_DELAY_LOAD -sRUNTIME_DEBUG -sSTACK_OVERFLOW_CHECK=2 -sTEXTDECODER=2 -sPTHREAD_POOL_SIZE_STRICT=2 -sASSERTIONS=2 -sINITIAL_MEMORY=$MAX_MEMORY -sPTHREAD_POOL_SIZE=$MAX_THREADS -sDISABLE_EXCEPTION_CATCHING=0 -sEXIT_RUNTIME=0 -sINVOKE_RUN=0 -sPOLYFILL=0 -sEXPORTED_FUNCTIONS=_malloc -sEXPORT_NAME=loadVosklet -sMALLOC=emmalloc -sEXPORTED_RUNTIME_METHODS=UTF8ToString,stringToUTF8OnStack -sENVIRONMENT=web,worker -I. -I$LIBARCHIVE/include -I$VOSK/src -L$LIBARCHIVE/lib -larchive -L$KALDI/src -l:online2/kaldi-online2.a -l:decoder/kaldi-decoder.a -l:ivector/kaldi-ivector.a -l:gmm/kaldi-gmm.a -l:tree/kaldi-tree.a -l:feat/kaldi-feat.a -l:cudamatrix/kaldi-cudamatrix.a -l:lat/kaldi-lat.a -l:lm/kaldi-lm.a -l:rnnlm/kaldi-rnnlm.a -l:hmm/kaldi-hmm.a -l:nnet3/kaldi-nnet3.a -l:transform/kaldi-transform.a -l:matrix/kaldi-matrix.a -l:fstext/kaldi-fstext.a -l:util/kaldi-util.a -l:base/kaldi-base.a -L$OPENFST/lib -l:libfst.a -l:libfstngram.a -L$CLAPACK_WASM -l:CBLAS/lib/cblas.a -l:CLAPACK-3.2.1/lapack.a -l:CLAPACK-3.2.1/libcblaswr.a -l:f2c_BLAS-3.8.0/blas.a -l:libf2c/libf2c.a -L$VOSK/src -l:vosk.a -lopfs.js -lembind -pthread -flto -msimd128 --embind-emit-tsd Vosklet.d.ts -fsanitize=undefined -fsanitize=address -fsanitize=leak -g3 --pre-js pre.js -o ../devel/Vosklet.js
elif [ $MODE = 1 ]; then elif [ $MODE = 1 ]; then
em++ -O3 global.cc genericModel.cc recognizer.cc bindings.cc -sWASMFS -sWASM_BIGINT -sSINGLE_FILE -sGZIP_EMBEDDINGS -sMODULARIZE -sEMBIND_STD_STRING_IS_UTF8 -sPTHREAD_POOL_DELAY_LOAD -sTEXTDECODER=2 -sPTHREAD_POOL_SIZE_STRICT=2 -sINITIAL_MEMORY=$MAX_MEMORY -sPTHREAD_POOL_SIZE=$MAX_THREADS -sPOLYFILL=0 -sEXIT_RUNTIME=0 -sINVOKE_RUN=0 -sSUPPORT_LONGJMP=0 -sEXPORTED_FUNCTIONS=_malloc -sEXPORT_NAME=loadVosklet -sMALLOC=emmalloc -sEXPORTED_RUNTIME_METHODS=UTF8ToString,stringToUTF8OnStack -sENVIRONMENT=web,worker -I. -I$LIBARCHIVE/include -I$VOSK/src -L$LIBARCHIVE/lib -larchive -L$KALDI/src -l:online2/kaldi-online2.a -l:decoder/kaldi-decoder.a -l:ivector/kaldi-ivector.a -l:gmm/kaldi-gmm.a -l:tree/kaldi-tree.a -l:feat/kaldi-feat.a -l:cudamatrix/kaldi-cudamatrix.a -l:lat/kaldi-lat.a -l:lm/kaldi-lm.a -l:rnnlm/kaldi-rnnlm.a -l:hmm/kaldi-hmm.a -l:nnet3/kaldi-nnet3.a -l:transform/kaldi-transform.a -l:matrix/kaldi-matrix.a -l:fstext/kaldi-fstext.a -l:util/kaldi-util.a -l:base/kaldi-base.a -L$OPENFST/lib -l:libfst.a -l:libfstngram.a -L$CLAPACK_WASM -l:CBLAS/lib/cblas.a -l:CLAPACK-3.2.1/lapack.a -l:CLAPACK-3.2.1/libcblaswr.a -l:f2c_BLAS-3.8.0/blas.a -l:libf2c/libf2c.a -L$VOSK/src -l:vosk.a -lopfs.js -lembind -pthread -flto -msimd128 --pre-js pre.js -o ../devel/Vosklet.js em++ -O3 global.cc genericModel.cc recognizer.cc bindings.cc -sWASMFS -sWASM_BIGINT -sSINGLE_FILE -sGZIP_EMBEDDINGS -sMODULARIZE -sEMBIND_STD_STRING_IS_UTF8 -sPTHREAD_POOL_DELAY_LOAD -sTEXTDECODER=2 -sPTHREAD_POOL_SIZE_STRICT=2 -sINITIAL_MEMORY=$MAX_MEMORY-sPTHREAD_POOL_SIZE=$MAX_THREADS -sPOLYFILL=0 -sEXIT_RUNTIME=0 -sINVOKE_RUN=0 -sSUPPORT_LONGJMP=0 -sEXPORTED_FUNCTIONS=_malloc -sEXPORT_NAME=loadVosklet -sMALLOC=emmalloc -sEXPORTED_RUNTIME_METHODS=UTF8ToString,stringToUTF8OnStack -sENVIRONMENT=web,worker -I. -I$LIBARCHIVE/include -I$VOSK/src -L$LIBARCHIVE/lib -larchive -L$KALDI/src -l:online2/kaldi-online2.a -l:decoder/kaldi-decoder.a -l:ivector/kaldi-ivector.a -l:gmm/kaldi-gmm.a -l:tree/kaldi-tree.a -l:feat/kaldi-feat.a -l:cudamatrix/kaldi-cudamatrix.a -l:lat/kaldi-lat.a -l:lm/kaldi-lm.a -l:rnnlm/kaldi-rnnlm.a -l:hmm/kaldi-hmm.a -l:nnet3/kaldi-nnet3.a -l:transform/kaldi-transform.a -l:matrix/kaldi-matrix.a -l:fstext/kaldi-fstext.a -l:util/kaldi-util.a -l:base/kaldi-base.a -L$OPENFST/lib -l:libfst.a -l:libfstngram.a -L$CLAPACK_WASM -l:CBLAS/lib/cblas.a -l:CLAPACK-3.2.1/lapack.a -l:CLAPACK-3.2.1/libcblaswr.a -l:f2c_BLAS-3.8.0/blas.a -l:libf2c/libf2c.a -L$VOSK/src -l:vosk.a -lopfs.js -lembind -pthread -flto -msimd128 --pre-js pre.js -o ../devel/Vosklet.js
else else
em++ -O0 global.cc genericModel.cc recognizer.cc bindings.cc -sWASMFS -sWASM_BIGINT -sSINGLE_FILE -sGZIP_EMBEDDINGS -sMODULARIZE -sEMBIND_STD_STRING_IS_UTF8 -sPTHREAD_POOL_DELAY_LOAD -sRUNTIME_DEBUG -sSTACK_OVERFLOW_CHECK=2 -sTEXTDECODER=2 -sPTHREAD_POOL_SIZE_STRICT=2 -sASSERTIONS=2 -sINITIAL_MEMORY=$MAX_MEMORY -sPTHREAD_POOL_SIZE=$MAX_THREADS -sDISABLE_EXCEPTION_CATCHING=0 -sEXIT_RUNTIME=0 -sINVOKE_RUN=0 -sPOLYFILL=0 -sEXPORTED_FUNCTIONS=_malloc -sEXPORT_NAME=loadVosklet -sMALLOC=emmalloc -sEXPORTED_RUNTIME_METHODS=UTF8ToString,stringToUTF8OnStack -sENVIRONMENT=web,worker -I. -I$LIBARCHIVE/include -I$VOSK/src -L$LIBARCHIVE/lib -larchive -L$KALDI/src -l:online2/kaldi-online2.a -l:decoder/kaldi-decoder.a -l:ivector/kaldi-ivector.a -l:gmm/kaldi-gmm.a -l:tree/kaldi-tree.a -l:feat/kaldi-feat.a -l:cudamatrix/kaldi-cudamatrix.a -l:lat/kaldi-lat.a -l:lm/kaldi-lm.a -l:rnnlm/kaldi-rnnlm.a -l:hmm/kaldi-hmm.a -l:nnet3/kaldi-nnet3.a -l:transform/kaldi-transform.a -l:matrix/kaldi-matrix.a -l:fstext/kaldi-fstext.a -l:util/kaldi-util.a -l:base/kaldi-base.a -L$OPENFST/lib -l:libfst.a -l:libfstngram.a -L$CLAPACK_WASM -l:CBLAS/lib/cblas.a -l:CLAPACK-3.2.1/lapack.a -l:CLAPACK-3.2.1/libcblaswr.a -l:f2c_BLAS-3.8.0/blas.a -l:libf2c/libf2c.a -L$VOSK/src -l:vosk.a -lopfs.js -lembind -pthread -flto -msimd128 --embind-emit-tsd Vosklet.d.ts -fsanitize=undefined -fsanitize=address -fsanitize=leak -g3 --pre-js pre.js -o ../devel/Vosklet.js em++ -O0 global.cc genericModel.cc recognizer.cc bindings.cc -sWASMFS -sWASM_BIGINT -sSINGLE_FILE -sGZIP_EMBEDDINGS -sMODULARIZE -sEMBIND_STD_STRING_IS_UTF8 -sPTHREAD_POOL_DELAY_LOAD -sRUNTIME_DEBUG -sSTACK_OVERFLOW_CHECK=2 -sTEXTDECODER=2 -sPTHREAD_POOL_SIZE_STRICT=2 -sASSERTIONS=2 -sINITIAL_MEMORY=$MAX_MEMORY -sPTHREAD_POOL_SIZE=$MAX_THREADS -sDISABLE_EXCEPTION_CATCHING=0 -sEXIT_RUNTIME=0 -sINVOKE_RUN=0 -sPOLYFILL=0 -sEXPORTED_FUNCTIONS=_malloc -sEXPORT_NAME=loadVosklet -sMALLOC=emmalloc -sEXPORTED_RUNTIME_METHODS=UTF8ToString,stringToUTF8OnStack -sENVIRONMENT=web,worker -I. -I$LIBARCHIVE/include -I$VOSK/src -L$LIBARCHIVE/lib -larchive -L$KALDI/src -l:online2/kaldi-online2.a -l:decoder/kaldi-decoder.a -l:ivector/kaldi-ivector.a -l:gmm/kaldi-gmm.a -l:tree/kaldi-tree.a -l:feat/kaldi-feat.a -l:cudamatrix/kaldi-cudamatrix.a -l:lat/kaldi-lat.a -l:lm/kaldi-lm.a -l:rnnlm/kaldi-rnnlm.a -l:hmm/kaldi-hmm.a -l:nnet3/kaldi-nnet3.a -l:transform/kaldi-transform.a -l:matrix/kaldi-matrix.a -l:fstext/kaldi-fstext.a -l:util/kaldi-util.a -l:base/kaldi-base.a -L$OPENFST/lib -l:libfst.a -l:libfstngram.a -L$CLAPACK_WASM -l:CBLAS/lib/cblas.a -l:CLAPACK-3.2.1/lapack.a -l:CLAPACK-3.2.1/libcblaswr.a -l:f2c_BLAS-3.8.0/blas.a -l:libf2c/libf2c.a -L$VOSK/src -l:vosk.a -lopfs.js -lembind -pthread -flto -msimd128 --embind-emit-tsd Vosklet.d.ts -fsanitize=undefined -fsanitize=address -fsanitize=leak -g3 --pre-js pre.js -o ../devel/Vosklet.js
fi fi

View File

@@ -5,30 +5,31 @@ void genericModel::load() {
thrd.addTask([this](){ thrd.addTask([this](){
if(normalMdl) mdl = vosk_model_new(storepath.c_str()); if(normalMdl) mdl = vosk_model_new(storepath.c_str());
else vosk_spk_model_new(storepath.c_str()); else vosk_spk_model_new(storepath.c_str());
if(normalMdl && std::get<0>(mdl) == nullptr || std::get<1>(mdl) == nullptr) fireEv(index, "Unable to load model for recognition"); emscripten_console_log("Loading finished!");
else fireEv(index, nullptr); if(normalMdl ? std::get<0>(mdl) == nullptr : std::get<1>(mdl) == nullptr) fireEv(index, "Unable to load model for recognition");
else fireEv(index, "0");
}); });
} }
bool genericModel::checkFiles() { bool genericModel::checkFiles() {
if(normalMdl) { if(normalMdl) {
return fs::exists(storepath + "/am/final.mdl", tank) && return fs::exists(storepath + "/am/final.mdl") &&
fs::exists(storepath + "/conf/mfcc.conf", tank) && fs::exists(storepath + "/conf/mfcc.conf") &&
fs::exists(storepath + "/conf/model.conf", tank) && fs::exists(storepath + "/conf/model.conf") &&
fs::exists(storepath + "/graph/phones/word_boundary.int", tank) && fs::exists(storepath + "/graph/phones/word_boundary.int") &&
fs::exists(storepath + "/graph/Gr.fst", tank) && fs::exists(storepath + "/graph/Gr.fst") &&
fs::exists(storepath + "/graph/HCLr.fst", tank) && fs::exists(storepath + "/graph/HCLr.fst") &&
fs::exists(storepath + "/graph/disambig_tid.int", tank) && fs::exists(storepath + "/graph/disambig_tid.int") &&
fs::exists(storepath + "/ivector/final.dubm", tank) && fs::exists(storepath + "/ivector/final.dubm") &&
fs::exists(storepath + "/ivector/final.ie", tank) && fs::exists(storepath + "/ivector/final.ie") &&
fs::exists(storepath + "/ivector/final.mat", tank) && fs::exists(storepath + "/ivector/final.mat") &&
fs::exists(storepath + "/ivector/global_cmvn.stats", tank) && fs::exists(storepath + "/ivector/global_cmvn.stats") &&
fs::exists(storepath + "/ivector/online_cmvn.conf", tank) && fs::exists(storepath + "/ivector/online_cmvn.conf") &&
fs::exists(storepath + "/ivector/splice.conf", tank); fs::exists(storepath + "/ivector/splice.conf");
} }
return fs::exists(storepath + "/mfcc.conf", tank) && return fs::exists(storepath + "/mfcc.conf") &&
fs::exists(storepath + "/final.ext.raw", tank) && fs::exists(storepath + "/final.ext.raw") &&
fs::exists(storepath + "/mean.vec", tank) && fs::exists(storepath + "/mean.vec") &&
fs::exists(storepath + "/transform.mat", tank); fs::exists(storepath + "/transform.mat");
} }
genericModel::~genericModel() { genericModel::~genericModel() {
archive_entry_free(entry); archive_entry_free(entry);
@@ -53,8 +54,8 @@ void genericModel::check() {
fireEv(index, "OPFS initialization failed"); fireEv(index, "OPFS initialization failed");
return; return;
} }
if(!checkFiles() && !fs::exists(storepath + "/id", tank)) { if(!checkFiles() && !fs::exists(storepath + "/id")) {
emscripten_console_log("Model is not available, fetching..."); emscripten_console_log("Model is not available or incorrect file, fetching...");
fireEv(index, "1"); fireEv(index, "1");
return; return;
} }
@@ -80,8 +81,8 @@ void genericModel::afterFetch() {
thrd.addTask([this](){ thrd.addTask([this](){
emscripten_console_log("Trying to extract..."); emscripten_console_log("Trying to extract...");
if(!extract()) { if(!extract()) {
fs::remove(storepath + ".tar",tank); fs::remove(storepath + ".tar");
fs::remove_all(storepath, tank); fs::remove_all(storepath);
emscripten_console_log("Unable to extract model"); emscripten_console_log("Unable to extract model");
fireEv(index, "Unable to extract model"); fireEv(index, "Unable to extract model");
return; return;
@@ -97,16 +98,16 @@ void genericModel::afterFetch() {
if(idFd == -1) { if(idFd == -1) {
emscripten_console_log("Unable to create ID file"); emscripten_console_log("Unable to create ID file");
fireEv(index, "Unable to create ID file"); fireEv(index, "Unable to create ID file");
fs::remove(storepath + ".tar",tank); fs::remove(storepath + ".tar");
fs::remove_all(storepath, tank); fs::remove_all(storepath);
return; return;
} }
emscripten_console_log("Writing ID..."); emscripten_console_log("Writing ID...");
if(write(idFd,id.data(), id.size()) == -1) { if(write(idFd,id.data(), id.size()) == -1) {
emscripten_console_log("Unable to write new ID"); emscripten_console_log("Unable to write new ID");
fireEv(index, "Unable to write new ID"); fireEv(index, "Unable to write new ID");
fs::remove(storepath + ".tar",tank); fs::remove(storepath + ".tar");
fs::remove_all(storepath, tank); fs::remove_all(storepath);
close(idFd); close(idFd);
return; return;
}; };
@@ -148,8 +149,8 @@ bool genericModel::extract() {
return false; return false;
} }
} }
fs::remove(storepath + "/README",tank); fs::remove(storepath + "/README");
fs::remove(storepath + ".tar" ,tank); fs::remove(storepath + ".tar" );
archive_read_free(src); archive_read_free(src);
return true; return true;
} }

View File

@@ -14,7 +14,7 @@ namespace fs = std::filesystem;
struct genericModel { struct genericModel {
bool normalMdl; bool normalMdl;
bool recognizerUsedThrd; bool recognizerUsedThrd{};
int index; int index;
std::string storepath; std::string storepath;
std::string id; std::string id;

View File

@@ -1,7 +1,6 @@
#include "global.h" #include "global.h"
pthread_t dstThrd{pthread_self()}; pthread_t dstThrd{pthread_self()};
std::error_code tank{};
bool OPFSOk{}; bool OPFSOk{};
bool OPFSTried{}; bool OPFSTried{};
ProxyingQueue glbQ{}; ProxyingQueue glbQ{};
@@ -19,9 +18,10 @@ void fireEv(int index, const char* content, const char* type) {
reusableThrd::reusableThrd() { reusableThrd::reusableThrd() {
std::thread thrd{[this](){ std::thread thrd{[this](){
while(!done.test(std::memory_order_relaxed)) { while(!done.test(std::memory_order_relaxed)) {
blocker.wait(done.test(std::memory_order_relaxed) || queue.empty(), std::memory_order_relaxed); blocker.wait(queue.empty() || done.test(std::memory_order_relaxed), std::memory_order_relaxed);
blocker.clear(std::memory_order_relaxed); blocker.clear(std::memory_order_relaxed);
while(!queue.empty()) { while(!queue.empty()) {
emscripten_console_log("==========Executing task===========");
queue.front()(); queue.front()();
queue.pop(); queue.pop();
} }
@@ -31,6 +31,8 @@ reusableThrd::reusableThrd() {
} }
void reusableThrd::addTask(std::function<void()>&& task) { void reusableThrd::addTask(std::function<void()>&& task) {
queue.emplace(task); queue.emplace(task);
// The following line will make -O3 magically work. Yes, I know, it is a god dang PRINT STATEMENT.
//emscripten_console_log("");
blocker.test_and_set(std::memory_order_relaxed); blocker.test_and_set(std::memory_order_relaxed);
blocker.notify_one(); blocker.notify_one();
} }

View File

@@ -1,14 +1,17 @@
#include "recognizer.h" #include "recognizer.h"
recognizer::recognizer(genericModel* model, float sampleRate, int index) : index(index) { recognizer::recognizer(genericModel* model, float sampleRate, int index) : index(index) {
emscripten_console_log("Recognizer constructor called...");
rec = vosk_recognizer_new(std::get<0>(model->mdl),sampleRate); rec = vosk_recognizer_new(std::get<0>(model->mdl),sampleRate);
finishConstruction(model, nullptr); finishConstruction(model, nullptr);
} }
recognizer::recognizer(genericModel* model, genericModel* spkMdl, float sampleRate, int index) { recognizer::recognizer(genericModel* model, genericModel* spkMdl, float sampleRate, int index) : index(index) {
emscripten_console_log("Recognizer constructor called...");
rec = vosk_recognizer_new_spk(std::get<0>(model->mdl), sampleRate, std::get<1>(spkMdl->mdl)); rec = vosk_recognizer_new_spk(std::get<0>(model->mdl), sampleRate, std::get<1>(spkMdl->mdl));
finishConstruction(model, spkMdl); finishConstruction(model, spkMdl);
} }
recognizer::recognizer(genericModel* model, const std::string& grm, float sampleRate, int index, int dummy) { recognizer::recognizer(genericModel* model, const std::string& grm, float sampleRate, int index, int dummy) : index(index) {
emscripten_console_log("Recognizer constructor called...");
rec = vosk_recognizer_new_grm(std::get<0>(model->mdl), sampleRate, grm.c_str()); rec = vosk_recognizer_new_grm(std::get<0>(model->mdl), sampleRate, grm.c_str());
finishConstruction(model, nullptr); finishConstruction(model, nullptr);
} }
@@ -20,12 +23,13 @@ recognizer::~recognizer() {
vosk_recognizer_free(rec); vosk_recognizer_free(rec);
free(dataPtr); free(dataPtr);
} }
void recognizer::finishConstruction(genericModel* model, genericModel* spkModel) { void recognizer:: finishConstruction(genericModel* model, genericModel* spkModel) {
if(rec == nullptr) { if(rec == nullptr) {
fireEv(index, "Unable to initialize recognizer"); fireEv(index, "Unable to initialize recognizer");
return; return;
} }
auto main {[this](){ auto main {[this](){
emscripten_console_log("Recognizer loaded!");
fireEv(index, "0"); fireEv(index, "0");
while(!done.test(std::memory_order_relaxed)) { while(!done.test(std::memory_order_relaxed)) {
controller.wait(!done.test(std::memory_order_relaxed), std::memory_order_relaxed); controller.wait(!done.test(std::memory_order_relaxed), std::memory_order_relaxed);
@@ -43,12 +47,16 @@ void recognizer::finishConstruction(genericModel* model, genericModel* spkModel)
if(!model->recognizerUsedThrd) { if(!model->recognizerUsedThrd) {
model->recognizerUsedThrd = true; model->recognizerUsedThrd = true;
model->thrd.addTask(main); model->thrd.addTask(main);
emscripten_console_log("Adding task to model thread...");
return; return;
} }
if(spkModel != nullptr && !spkModel->recognizerUsedThrd) { if(spkModel != nullptr && !spkModel->recognizerUsedThrd) {
spkModel->recognizerUsedThrd = true; spkModel->recognizerUsedThrd = true;
spkModel->thrd.addTask(main); spkModel->thrd.addTask(main);
emscripten_console_log("Adding task to speaker model thread...");
return;
} }
emscripten_console_log("New recognizer thread");
std::thread t{main}; std::thread t{main};
t.detach(); t.detach();
} }