More testing

devel/Vosklet.worker.js

@@ -1 +1,163 @@
-"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['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;

devel/index.html

@@ -3,8 +3,20 @@
   <head>
     <script src="Vosklet.js"></script>
     <script>
-      window.onload = async () => {
-        window.ctx = new AudioContext()
+      async function func() {
+        let processorURL = URL.createObjectURL(new Blob(['(',
+          (() => {
+            registerProcessor('processor', class extends AudioWorkletProcessor {
+              process(inputs, outputs) {
+                for (let channel = 0; channel < outputs[0].length; ++channel) {
+                  outputs[0][channel].set(inputs[0][channel])
+                }
+                return true
+              }
+            })
+          }).toString(),
+        ')()'], { type : "text/javascript" }))
+        window.ctx = new AudioContext({sampleRate : 16000})
         let micNode = ctx.createMediaStreamSource(await navigator.mediaDevices.getUserMedia({
           video: false,
           audio: {
@@ -14,17 +26,26 @@
             sampleRate: 16000
           },
         }))
-        window.module = await loadVosklet()
+        await ctx.audioWorklet.addModule(processorURL)
+        window.node = new AudioWorkletNode(ctx, "processor", { channelCountMode: "explicit", numberOfInputs : 1, numberOfOutputs : 1})
+        micNode.connect(node).connect(ctx.destination)
+        ctx.resume()
+      }
+      window.onload = async () => {
+        /*window.module = await loadVosklet()
         window.model = await module.makeModel("../usage/en-model.tgz","model","ID")
-        window.recognizer = await module.makeRecognizer(model, ctx.sampleRate)
+        window.recognizer = await module.makeRecognizer(model, 16000)
         recognizer.addEventListener("result", e => {
           console.log("Result: ", e.detail)
         })
         recognizer.addEventListener("partialResult", e => {
           console.log("Partial result: ", e.detail)
-        })
-        //let recNode = recognizer.getNode(ctx);
+        })*/
+        //let recNode = await recognizer.getNode(ctx)
+        //recNode.connect(ctx.destination)
+        //micNode.connect(recNode)
       }
     </script>
+    <button onclick="func()">Just another button</button>
   </head>
 </html>

devel/test.sh

@@ -28,12 +28,12 @@ fi
 . $EMSDK/emsdk_env.sh &&
 
 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" &&
 if [ $MODE = 0 ]; then
   em++ -O0 link.cc genericModel.cc recognizer.cc bindings.cc -sWASMFS -sWASM_BIGINT -sSINGLE_FILE -sMODULARIZE -sEMBIND_STD_STRING_IS_UTF8 -sPTHREAD_POOL_DELAY_LOAD -sRUNTIME_DEBUG -sSTACK_OVERFLOW_CHECK=2 -sTEXTDECODER=2 -sPTHREAD_POOL_SIZE_STRICT=2 -sASSERTIONS=2 -sINITIAL_MEMORY=$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 -lembind -pthread -flto -msimd128 --embind-emit-tsd Vosklet.d.ts -fsanitize=undefined -fsanitize=address -fsanitize=leak -msimd128 -mreference-types -mnontrapping-fptoint -mextended-const -msign-ext -g3 --pre-js pre.js -o ../devel/Vosklet.js 
 elif [ $MODE = 1 ]; then
   em++ -O3 link.cc genericModel.cc recognizer.cc bindings.cc -sWASMFS -sWASM_BIGINT -sSINGLE_FILE -sMODULARIZE -sEMBIND_STD_STRING_IS_UTF8 -sPTHREAD_POOL_DELAY_LOAD -sTEXTDECODER=2 -sPTHREAD_POOL_SIZE_STRICT=2 -sINITIAL_MEMORY=$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 -lembind -pthread -flto -msimd128 -mreference-types -mnontrapping-fptoint -mextended-const -msign-ext --pre-js pre.js -o ../devel/Vosklet.js
 else
-  em++ -O3 link.cc genericModel.cc recognizer.cc bindings.cc -sWASMFS -sWASM_BIGINT -sSINGLE_FILE -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$VOSK/src -l:vosk.a -lembind -pthread -flto -msimd128 -mreference-types -mnontrapping-fptoint -mextended-const -msign-ext -g3 --pre-js pre.js -o ../devel/Vosklet.js
+  em++ -O0 link.cc genericModel.cc recognizer.cc bindings.cc -sWASMFS -sWASM_BIGINT -sSINGLE_FILE -sMODULARIZE -sEMBIND_STD_STRING_IS_UTF8 -sPTHREAD_POOL_DELAY_LOAD -sRUNTIME_DEBUG -sSTACK_OVERFLOW_CHECK=2 -sTEXTDECODER=2 -sPTHREAD_POOL_SIZE_STRICT=2 -sASSERTIONS=2 -sINITIAL_MEMORY=$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 -lembind -pthread -flto -msimd128 -msimd128 -mreference-types -mnontrapping-fptoint -mextended-const -msign-ext -g3 --pre-js pre.js -o ../devel/Vosklet.js 
 fi

src/bindings.cc

@@ -21,6 +21,7 @@ EMSCRIPTEN_BINDINGS() {
   .constructor<int, float, genericModel*>(allow_raw_pointers())
   .constructor<int, float, genericModel*, genericModel*>(allow_raw_pointers())
   .constructor<int, float, genericModel*, std::string, int>(allow_raw_pointers())
+  .function("acceptWaveform", &recognizer::acceptWaveform, allow_raw_pointers())
   .function("reset", &recognizer::reset, allow_raw_pointers())
   .function("setEndpointerMode", &recognizer::setEndpointerMode, allow_raw_pointers())
   .function("setEndpointerDelays", &recognizer::setEndpointerDelays, allow_raw_pointers())

src/genericModel.cc

@@ -49,7 +49,7 @@ void genericModel::extractAndLoad(int tarStart, int tarSize) {
     archive_read_free(src);
     if(normalMdl) mdl = vosk_model_new(storepath.c_str());
     else vosk_spk_model_new(storepath.c_str());
-    emscripten_console_log("Loading finished!");
+    emscripten_console_log("Model loaded!");
     if(normalMdl ? std::get<0>(mdl) == nullptr : std::get<1>(mdl) == nullptr) fireEv(index, "Unable to load model for recognition");
     else fireEv(index, "0");
   };

src/link.cc

@@ -1,6 +1,6 @@
 #include "link.h"
 
-pthread_t dstThrd{pthread_self()};
+auto dstThrd = pthread_self();
 ProxyingQueue glbQ{};
 void fireEv(int index, const char* content, const char* type) {
   auto proxy{[index, content, type](){
@@ -11,12 +11,3 @@ void fireEv(int index, const char* content, const char* type) {
   if(dstThrd == pthread_self()) proxy();
   else glbQ.proxySync(dstThrd, proxy);
 }
-void fireEv(int index, std::atomic_int* state, float* dataBuf) {
-  auto proxy{[index, state, dataBuf](){
-    EM_ASM({
-      objs[$0].dispatchEvent(new CustomEvent("0", { "detail" : $1 + "," + $2}));
-    }, index, state, dataBuf);
-  }};
-  if(dstThrd == pthread_self()) proxy();
-  else glbQ.proxySync(dstThrd, proxy);
-}

src/link.h

@@ -1,7 +1,8 @@
 #pragma once
+#include <thread>
+
 #include <emscripten/em_asm.h>
 #include <emscripten/proxying.h>
 using namespace emscripten;
 
 void fireEv(int index, const char* content, const char* type = nullptr); // Normal
-void fireEv(int index, std::atomic_int* state, float* dataBuf); // For recognizer's success initialization

src/pre.js

@@ -29,13 +29,9 @@ class genericModel extends EventTarget {
     let mdl = new genericModel(url, storepath, id, normalMdl)
     let result = new Promise((resolve, reject) => {
       mdl.addEventListener("0", ev => {
-        switch(ev.detail) {
-          case "0":
-            return resolve(mdl)
-          default:
+        if(ev.detail === "0") return resolve(mdl)
         mdl.delete()
         reject(ev.detail)
-        }
       }, { once : true })
     })
     let tar
@@ -93,12 +89,7 @@ class Recognizer extends EventTarget {
     let rec = new Recognizer()
     let result = new Promise((resolve, reject) => {
       rec.addEventListener("0", ev => {
-        if(ev.detail.indexOf(",") !== -1) {
-          let loadInfo = ev.detail.split(",")
-          rec.state = Module.HEAP32.subarray(parseInt(loadInfo[0]), parseInt(loadInfo[0]) + 1) // State is an array with 1 element, there is no other way to get a reference to a single element
-          rec.dataBuf = Module.HEAPF32.subarray(parseInt(loadInfo[1]), parseInt(loadInfo[1]) + 128)
-          return resolve(rec)
-        }
+        if(ev.detail === "0") return resolve(rec)
         rec.delete()
         reject(ev.detail)
       }, { once : true })
@@ -115,16 +106,15 @@ class Recognizer extends EventTarget {
     }
     return result
   } 
-  async getNode(ctx, channelIndex = 0) {
+  async getNode(ctx) {
     if(typeof this.node === "undefined") {
-      await ctx.audioWorklet.addModule(processorUrl)
-      this.node = new AudioWorkletNode(ctx, 'VoskletProcessor', { channelCountMode: "max", numberOfInputs: 1, numberOfOutputs: 0, processorOptions: { dataBuf: this.dataBuf, state: this.state, channel: channelIndex }})
+      await ctx.audioWorklet.addModule("../src/processor.js", { credentials : "omit"})
+      this.node = new AudioWorkletNode(ctx, 'VoskletProcessor', { channelCountMode: "explicit", channelCount: 1, numberOfInputs: 1, numberOfOutputs: 1, processorOptions: { dataBuf: this.dataBuf, state: this.state }})
     }
     return this.node
   }
-  recognize(buf, channelIndex = 0) {
-    Module.HEAPF32.set(buf.getChannelData(channelIndex).subarray(0, 512), this.ptr)
-    this.obj.acceptWaveForm()
+  recognize(buf) {
+    Module.HEAPF32.set(buf.getChannelData(0).subarray(0, 512), this.ptr)
   }
   delete() {
     if (this.obj) this.obj.delete()
@@ -158,7 +148,7 @@ Module.makeRecognizerWithSpkModel = (model, sampleRate, spkModel) => {
 Module.makeRecognizerWithGrm = (model, sampleRate, grammar) => {
   return Recognizer._init(model.obj, sampleRate, 3, grammar, null)
 } 
-let processorUrl = URL.createObjectURL(new Blob(['(',
+/*let processorURL = URL.createObjectURL(new Blob(['(',
   (() => {
     registerProcessor("VoskletProcessor", class extends AudioWorkletProcessor {
       constructor(options) {
@@ -167,15 +157,14 @@ let processorUrl = URL.createObjectURL(new Blob(['(',
         this.state = options.processorOptions.state
       }
       process(inputs, outputs, params) {
-        Atomics.wait(state, 0)
-        inputs.copyFromChannel(this.dataBuf, this.channelIndex)
-        
+        while(state[0])
+        inputs.copyFromChannel(this.dataBuf, 0)
         return true
       }
     })
-  }).toString()
-, ')()'], {type : "text/javascript"}))
-/*let pthreadUrl = URL.createObjectURL(new Blob(['(',
+  }).toString(),
+  ')()'], {type : "text/javascript"}))
+let pthreadURL = URL.createObjectURL(new Blob(['(',
   (() => {
     { PTHREAD_SCRIPT }
   }).toString()

src/recognizer.cc

@@ -1,39 +1,36 @@
 #include "recognizer.h" 
 
-recognizer::recognizer(int index, float sampleRate, genericModel* model) : index{index}, sampleRate{sampleRate}, rec{vosk_recognizer_new(std::get<0>(model->mdl),sampleRate)} {
+recognizer::recognizer(int index, float sampleRate, genericModel* model) : index{index}, rec{vosk_recognizer_new(std::get<0>(model->mdl),sampleRate)} {
   finishConstruction(model);
 }
-recognizer::recognizer(int index, float sampleRate, genericModel* model, genericModel* spkModel) : index(index), sampleRate{sampleRate}, rec{vosk_recognizer_new_spk(std::get<0>(model->mdl), sampleRate, std::get<1>(spkModel->mdl))} {
+recognizer::recognizer(int index, float sampleRate, genericModel* model, genericModel* spkModel) : index{index}, rec{vosk_recognizer_new_spk(std::get<0>(model->mdl), sampleRate, std::get<1>(spkModel->mdl))} {
   finishConstruction(model, spkModel);
 }
-recognizer::recognizer(int index, float sampleRate, genericModel* model, const std::string& grm, int dummy) : index{index}, sampleRate{sampleRate}, rec{vosk_recognizer_new_grm(std::get<0>(model->mdl), sampleRate, grm.c_str())} {
+recognizer::recognizer(int index, float sampleRate, genericModel* model, const std::string& grm, int dummy) : index{index}, rec{vosk_recognizer_new_grm(std::get<0>(model->mdl), sampleRate, grm.c_str())} {
   finishConstruction(model);
 }
 recognizer::~recognizer() {
   done = true;
   vosk_recognizer_free(rec);
 }
-void recognizer::reset() {
-  vosk_recognizer_reset(rec);
-}
 void recognizer::finishConstruction(genericModel* model, genericModel* spkModel) {
   if(rec == nullptr) {
     fireEv(index, "Unable to initialize recognizer");
     return;
   }
   auto main {[this](){
-    emscripten_console_log("Recognizer loaded!");
-    fireEv(index, &state, dataBuf);
     while(!done) {
-      switch(vosk_recognizer_accept_waveform_f(rec, dataBuf, 512)) {
+      while(!dataQ.empty()) {
+        switch(vosk_recognizer_accept_waveform_f(rec, dataQ.front().data, dataQ.front().len)) {
         case 0:
           fireEv(index, vosk_recognizer_result(rec), "result");
           break;
         case 1:
           fireEv(index, vosk_recognizer_partial_result(rec), "partialResult");
         }
-      state = 0;
-      state.wait(0, std::memory_order_relaxed);
+        free(dataQ.front().data);
+        dataQ.pop();
+      }
     }
   }};
   if(!model->resourceUsed) {
@@ -54,6 +51,12 @@ void recognizer::finishConstruction(genericModel* model, genericModel* spkModel)
   std::thread t{main};
   t.detach();
 }
+void recognizer::acceptWaveform(int start, int len) {
+  dataQ.emplace(start, len);
+}
+void recognizer::reset() {
+  vosk_recognizer_reset(rec);
+}
 void recognizer::setEndpointerMode(VoskEndpointerMode mode) {
   vosk_recognizer_set_endpointer_mode(rec, mode);
 }

src/recognizer.h

@@ -1,12 +1,16 @@
 #pragma once
 #include "genericModel.h"
-#include <condition_variable>
 
+#include <condition_variable>
+#include <queue>
+struct audioData {
+  float* data;
+  int len;
+  audioData(int start, int len) : data{reinterpret_cast<float*>(start)}, len{len} {}
+};
 struct recognizer {
   std::atomic_bool done;
-  std::atomic_int state; // 0: Copying data from JS, 1: Processing from C++
-  float dataBuf[128];
-  float sampleRate;
+  std::queue<audioData> dataQ{};
   int index;
   VoskRecognizer* rec;
   recognizer(int index, float sampleRate, genericModel* model);
@@ -14,6 +18,7 @@ struct recognizer {
   recognizer(int index, float sampleRate, genericModel* model, const std::string& grm, int dummy);
   ~recognizer();
   void finishConstruction(genericModel* model, genericModel* spkModel = nullptr);
+  void acceptWaveform(int start, int len);
   void reset();
   void setEndpointerMode(VoskEndpointerMode mode);
   void setEndpointerDelays(float tStartMax, float tEnd, float tMax);

usage/README.md

@@ -21,8 +21,7 @@
 ## ```Recognizer``` object 
 | Function/Object | Description |
 |---|---|
-| ```Promise<AudioWorkletNode> getNode(ctx: AudioContext, channelIndex = 0: int)``` | Get a node that process audio. It has 1 input and no output, **channelIndex** must point to a 16-bit mono channel of the input |
-| ```recognize(buf: AudioBuffer, channelIndex = 0: int)``` | Recognize an AudioBuffer, usually from something like ```BaseAudioContext.decodeAudioData()```, **channelIndex** must point to a 16-bit mono channel of **buf**
+| ```acceptWaveform(buf: AudioBuffer || Float32Array )``` | Recognize an AudioBuffer, usually from something like ```BaseAudioContext.decodeAudioData()```, **channelIndex** must point to a 16-bit mono channel of **buf**
 | ```setPartialWords(partialWords: bool)``` | See Vosk's description (default: false) |
 | ```setWords(words: bool)``` | See Vosk's description (default: false) |
 | ```setNLSML(nlsml: bool)``` | See Vosk's description (default: false) |