Node.js package (#260)

* npm : preparing infra for node package

* npm : package infra ready

* npm : initial version ready

* npm : change name to whisper.cpp

whisper.js is taken

bindings/javascript/CMakeLists.txt

@@ -20,15 +20,22 @@ if (WHISPER_WASM_SINGLE_FILE)
         ${CMAKE_BINARY_DIR}/bin/libwhisper.js
         ${CMAKE_CURRENT_SOURCE_DIR}/whisper.js
         )
+
+    add_custom_command(
+        TARGET ${TARGET} POST_BUILD
+        COMMAND ${CMAKE_COMMAND} -E copy
+        ${CMAKE_BINARY_DIR}/bin/libwhisper.worker.js
+        ${CMAKE_CURRENT_SOURCE_DIR}/libwhisper.worker.js
+        )
 endif()
 
 set_target_properties(${TARGET} PROPERTIES LINK_FLAGS " \
     --bind \
+    -s MODULARIZE=1 \
+    -s EXPORT_NAME=\"'whisper_factory'\" \
+    -s FORCE_FILESYSTEM=1 \
     -s USE_PTHREADS=1 \
     -s PTHREAD_POOL_SIZE=8 \
-    -s INITIAL_MEMORY=1610612736 \
-    -s TOTAL_MEMORY=1610612736 \
-    -s FORCE_FILESYSTEM=1 \
-    -s EXPORTED_RUNTIME_METHODS=\"['print', 'printErr', 'ccall', 'cwrap']\" \
+    -s ALLOW_MEMORY_GROWTH=1 \
     ${EXTRA_FLAGS} \
     ")

bindings/javascript/README.md

@@ -0,0 +1,5 @@
+# whisper.cpp
+
+Node.js package for Whisper speech recognition
+
+For sample usage check [tests/test-whisper.js](/tests/test-whisper.js)

bindings/javascript/emscripten.cpp

@@ -1,63 +1,48 @@
+//
+// This is the Javascript API of whisper.cpp
+//
+// Very crude at the moment.
+// Feel free to contribute and make this better!
+//
+// See the tests/test-whisper.js for sample usage
+//
+
 #include "whisper.h"
 
 #include <emscripten.h>
 #include <emscripten/bind.h>
 
-#include <vector>
 #include <thread>
+#include <vector>
 
-std::thread g_worker;
-
-std::vector<struct whisper_context *> g_contexts(4, nullptr);
+struct whisper_context * g_context;
 
 EMSCRIPTEN_BINDINGS(whisper) {
     emscripten::function("init", emscripten::optional_override([](const std::string & path_model) {
-        if (g_worker.joinable()) {
-            g_worker.join();
-        }
-
-        for (size_t i = 0; i < g_contexts.size(); ++i) {
-            if (g_contexts[i] == nullptr) {
-                g_contexts[i] = whisper_init(path_model.c_str());
-                if (g_contexts[i] != nullptr) {
-                    return i + 1;
-                } else {
-                    return (size_t) 0;
-                }
+        if (g_context == nullptr) {
+            g_context = whisper_init(path_model.c_str());
+            if (g_context != nullptr) {
+                return true;
+            } else {
+                return false;
             }
         }
 
-        return (size_t) 0;
+        return false;
     }));
 
-    emscripten::function("free", emscripten::optional_override([](size_t index) {
-        if (g_worker.joinable()) {
-            g_worker.join();
-        }
-
-        --index;
-
-        if (index < g_contexts.size()) {
-            whisper_free(g_contexts[index]);
-            g_contexts[index] = nullptr;
+    emscripten::function("free", emscripten::optional_override([]() {
+        if (g_context) {
+            whisper_free(g_context);
+            g_context = nullptr;
         }
     }));
 
-    emscripten::function("full_default", emscripten::optional_override([](size_t index, const emscripten::val & audio, const std::string & lang, bool translate) {
-        if (g_worker.joinable()) {
-            g_worker.join();
-        }
-
-        --index;
-
-        if (index >= g_contexts.size()) {
+    emscripten::function("full_default", emscripten::optional_override([](const emscripten::val & audio, const std::string & lang, bool translate) {
+        if (g_context == nullptr) {
             return -1;
         }
 
-        if (g_contexts[index] == nullptr) {
-            return -2;
-        }
-
         struct whisper_full_params params = whisper_full_default_params(whisper_sampling_strategy::WHISPER_SAMPLING_GREEDY);
 
         params.print_realtime   = true;
@@ -65,7 +50,7 @@ EMSCRIPTEN_BINDINGS(whisper) {
         params.print_timestamps = true;
         params.print_special    = false;
         params.translate        = translate;
-        params.language         = whisper_is_multilingual(g_contexts[index]) ? lang.c_str() : "en";
+        params.language         = whisper_is_multilingual(g_context) ? lang.c_str() : "en";
         params.n_threads        = std::min(8, (int) std::thread::hardware_concurrency());
         params.offset_ms        = 0;
 
@@ -82,9 +67,11 @@ EMSCRIPTEN_BINDINGS(whisper) {
 
         // print system information
         {
+            printf("\n");
             printf("system_info: n_threads = %d / %d | %s\n",
                     params.n_threads, std::thread::hardware_concurrency(), whisper_print_system_info());
 
+            printf("\n");
             printf("%s: processing %d samples, %.1f sec, %d threads, %d processors, lang = %s, task = %s ...\n",
                     __func__, int(pcmf32.size()), float(pcmf32.size())/WHISPER_SAMPLE_RATE,
                     params.n_threads, 1,
@@ -94,13 +81,11 @@ EMSCRIPTEN_BINDINGS(whisper) {
             printf("\n");
         }
 
-        // run the worker
+        // run whisper
         {
-            g_worker = std::thread([index, params, pcmf32 = std::move(pcmf32)]() {
-                whisper_reset_timings(g_contexts[index]);
-                whisper_full(g_contexts[index], params, pcmf32.data(), pcmf32.size());
-                whisper_print_timings(g_contexts[index]);
-            });
+            whisper_reset_timings(g_context);
+            whisper_full(g_context, params, pcmf32.data(), pcmf32.size());
+            whisper_print_timings(g_context);
         }
 
         return 0;

bindings/javascript/libwhisper.worker.js

@@ -0,0 +1 @@
+"use strict";var Module={};var ENVIRONMENT_IS_NODE=typeof process=="object"&&typeof process.versions=="object"&&typeof process.versions.node=="string";if(ENVIRONMENT_IS_NODE){var nodeWorkerThreads=require("worker_threads");var parentPort=nodeWorkerThreads.parentPort;parentPort.on("message",data=>onmessage({data:data}));var fs=require("fs");Object.assign(global,{self:global,require:require,Module:Module,location:{href:__filename},Worker:nodeWorkerThreads.Worker,importScripts:function(f){(0,eval)(fs.readFileSync(f,"utf8")+"//# sourceURL="+f)},postMessage:function(msg){parentPort.postMessage(msg)},performance:global.performance||{now:function(){return Date.now()}}})}var initializedJS=false;var pendingNotifiedProxyingQueues=[];function threadPrintErr(){var text=Array.prototype.slice.call(arguments).join(" ");if(ENVIRONMENT_IS_NODE){fs.writeSync(2,text+"\n");return}console.error(text)}function threadAlert(){var text=Array.prototype.slice.call(arguments).join(" ");postMessage({cmd:"alert",text:text,threadId:Module["_pthread_self"]()})}var err=threadPrintErr;self.alert=threadAlert;Module["instantiateWasm"]=(info,receiveInstance)=>{var instance=new WebAssembly.Instance(Module["wasmModule"],info);receiveInstance(instance);Module["wasmModule"]=null;return instance.exports};self.onunhandledrejection=e=>{throw e.reason??e};self.onmessage=e=>{try{if(e.data.cmd==="load"){Module["wasmModule"]=e.data.wasmModule;for(const handler of e.data.handlers){Module[handler]=function(){postMessage({cmd:"callHandler",handler:handler,args:[...arguments]})}}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)}whisper_factory(Module).then(function(instance){Module=instance})}else if(e.data.cmd==="run"){Module["__performance_now_clock_drift"]=performance.now()-e.data.time;Module["__emscripten_thread_init"](e.data.pthread_ptr,0,0,1);Module["establishStackSpace"]();Module["PThread"].receiveObjectTransfer(e.data);Module["PThread"].threadInitTLS();if(!initializedJS){Module["__embind_initialize_bindings"]();pendingNotifiedProxyingQueues.forEach(queue=>{Module["executeNotifiedProxyingQueue"](queue)});pendingNotifiedProxyingQueues=[];initializedJS=true}try{Module["invokeEntryPoint"](e.data.start_routine,e.data.arg)}catch(ex){if(ex!="unwind"){if(ex instanceof Module["ExitStatus"]){if(Module["keepRuntimeAlive"]()){}else{Module["__emscripten_thread_exit"](ex.status)}}else{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==="processProxyingQueue"){if(initializedJS){Module["executeNotifiedProxyingQueue"](e.data.queue)}else{pendingNotifiedProxyingQueues.push(e.data.queue)}}else if(e.data.cmd){err("worker.js received unknown command "+e.data.cmd);err(e.data)}}catch(ex){if(Module["__emscripten_thread_crashed"]){Module["__emscripten_thread_crashed"]()}throw ex}};

bindings/javascript/package-tmpl.json

@@ -0,0 +1,26 @@
+{
+  "name": "whisper.cpp",
+  "version": "@PROJECT_VERSION@",
+  "description": "Whisper speech recognition",
+  "main": "whisper.js",
+  "scripts": {
+    "test": "echo \"todo: add tests\" && exit 0"
+  },
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/ggerganov/whisper.cpp"
+  },
+  "keywords": [
+    "openai",
+    "whisper",
+    "speech-to-text",
+    "speech-recognition",
+    "transformer"
+  ],
+  "author": "Georgi Gerganov",
+  "license": "MIT",
+  "bugs": {
+    "url": "https://github.com/ggerganov/whisper.cpp/issues"
+  },
+  "homepage": "https://github.com/ggerganov/whisper.cpp#readme"
+}

bindings/javascript/package.json

@@ -0,0 +1,26 @@
+{
+  "name": "whisper.cpp",
+  "version": "1.0.0",
+  "description": "Whisper speech recognition",
+  "main": "whisper.js",
+  "scripts": {
+    "test": "echo \"todo: add tests\" && exit 0"
+  },
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/ggerganov/whisper.cpp"
+  },
+  "keywords": [
+    "openai",
+    "whisper",
+    "speech-to-text",
+    "speech-recognition",
+    "transformer"
+  ],
+  "author": "Georgi Gerganov",
+  "license": "MIT",
+  "bugs": {
+    "url": "https://github.com/ggerganov/whisper.cpp/issues"
+  },
+  "homepage": "https://github.com/ggerganov/whisper.cpp#readme"
+}