stream.wasm : add web-based real-time transcription (#112)

examples/CMakeLists.txt

@@ -20,6 +20,7 @@ include_directories(${CMAKE_CURRENT_SOURCE_DIR})
 
 if (EMSCRIPTEN)
     add_subdirectory(whisper.wasm)
+    add_subdirectory(stream.wasm)
     add_subdirectory(talk.wasm)
 else()
     add_subdirectory(main)

examples/helpers.js

@@ -19,6 +19,12 @@ var printTextarea = (function() {
     };
 })();
 
+async function clearCache() {
+    if (confirm('Are you sure you want to clear the cache?\nAll the models will be downloaded again.')) {
+        indexedDB.deleteDatabase(dbName);
+    }
+}
+
 // fetch a remote file from remote URL using the Fetch API
 async function fetchRemote(url, cbProgress, cbPrint) {
     cbPrint('fetchRemote: downloading with fetch()...');

examples/stream.wasm/CMakeLists.txt

@@ -0,0 +1,47 @@
+#
+# libstream
+#
+
+set(TARGET libstream)
+
+add_executable(${TARGET}
+    emscripten.cpp
+    )
+
+target_link_libraries(${TARGET} PRIVATE
+    whisper
+    )
+
+unset(EXTRA_FLAGS)
+
+if (WHISPER_WASM_SINGLE_FILE)
+    set(EXTRA_FLAGS "-s SINGLE_FILE=1")
+    message(STATUS "Embedding WASM inside stream.js")
+
+    add_custom_command(
+        TARGET ${TARGET} POST_BUILD
+        COMMAND ${CMAKE_COMMAND} -E copy
+        ${CMAKE_BINARY_DIR}/bin/libstream.js
+        ${CMAKE_RUNTIME_OUTPUT_DIRECTORY}/stream.wasm/stream.js
+        )
+endif()
+
+set_target_properties(${TARGET} PROPERTIES LINK_FLAGS " \
+    --bind \
+    -s USE_PTHREADS=1 \
+    -s PTHREAD_POOL_SIZE=8 \
+    -s INITIAL_MEMORY=1024MB \
+    -s TOTAL_MEMORY=1024MB \
+    -s FORCE_FILESYSTEM=1 \
+    -s EXPORTED_RUNTIME_METHODS=\"['print', 'printErr', 'ccall', 'cwrap']\" \
+    ${EXTRA_FLAGS} \
+    ")
+
+#
+# stream.wasm
+#
+
+set(TARGET stream.wasm)
+
+configure_file(${CMAKE_CURRENT_SOURCE_DIR}/index-tmpl.html  ${CMAKE_RUNTIME_OUTPUT_DIRECTORY}/${TARGET}/index.html @ONLY)
+configure_file(${CMAKE_CURRENT_SOURCE_DIR}/../helpers.js    ${CMAKE_RUNTIME_OUTPUT_DIRECTORY}/${TARGET}/helpers.js @ONLY)

examples/stream.wasm/README.md

@@ -0,0 +1,20 @@
+# stream.wasm
+
+Real-time transcription in the browser using WebAssembly
+
+Online demo: https://whisper.ggerganov.com/stream/
+
+## Build instructions
+
+```bash
+# build using Emscripten (v3.1.2)
+git clone https://github.com/ggerganov/whisper.cpp
+cd whisper.cpp
+mkdir build-em && cd build-em
+emcmake cmake ..
+make -j
+
+# copy the produced page to your HTTP path
+cp bin/stream.wasm/*       /path/to/html/
+cp bin/libstream.worker.js /path/to/html/
+```

examples/stream.wasm/emscripten.cpp

@@ -0,0 +1,213 @@
+#include "ggml.h"
+#include "whisper.h"
+
+#include <emscripten.h>
+#include <emscripten/bind.h>
+
+#include <atomic>
+#include <cmath>
+#include <mutex>
+#include <string>
+#include <thread>
+#include <vector>
+
+constexpr int N_THREAD = 8;
+
+std::vector<struct whisper_context *> g_contexts(4, nullptr);
+
+std::mutex g_mutex;
+std::thread g_worker;
+
+std::atomic<bool> g_running(false);
+
+std::string g_status        = "";
+std::string g_status_forced = "";
+std::string g_transcribed   = "";
+
+std::vector<float> g_pcmf32;
+
+void stream_set_status(const std::string & status) {
+    std::lock_guard<std::mutex> lock(g_mutex);
+    g_status = status;
+}
+
+void stream_main(size_t index) {
+    stream_set_status("loading data ...");
+
+    struct whisper_full_params wparams = whisper_full_default_params(whisper_sampling_strategy::WHISPER_SAMPLING_GREEDY);
+
+    wparams.n_threads        = std::min(N_THREAD, (int) std::thread::hardware_concurrency());
+    wparams.offset_ms        = 0;
+    wparams.translate        = false;
+    wparams.no_context       = true;
+    wparams.single_segment   = true;
+    wparams.print_realtime   = false;
+    wparams.print_progress   = false;
+    wparams.print_timestamps = true;
+    wparams.print_special    = false;
+
+    wparams.max_tokens       = 32;
+    wparams.audio_ctx        = 768; // partial encoder context for better performance
+
+    wparams.language         = "en";
+
+    printf("stream: using %d threads\n", N_THREAD);
+
+    std::vector<float> pcmf32;
+
+    // whisper context
+    auto & ctx = g_contexts[index];
+
+    // 5 seconds interval
+    const int64_t window_samples = 5*WHISPER_SAMPLE_RATE;
+
+    while (g_running) {
+        stream_set_status("waiting for audio ...");
+
+        {
+            std::unique_lock<std::mutex> lock(g_mutex);
+
+            if (g_pcmf32.size() < 1024) {
+                lock.unlock();
+
+                std::this_thread::sleep_for(std::chrono::milliseconds(10));
+
+                continue;
+            }
+
+            pcmf32 = std::vector<float>(g_pcmf32.end() - std::min((int64_t) g_pcmf32.size(), window_samples), g_pcmf32.end());
+            g_pcmf32.clear();
+        }
+
+        {
+            const auto t_start = std::chrono::high_resolution_clock::now();
+
+            stream_set_status("running whisper ...");
+
+            int ret = whisper_full(ctx, wparams, pcmf32.data(), pcmf32.size());
+            if (ret != 0) {
+                printf("whisper_full() failed: %d\n", ret);
+                break;
+            }
+
+            const auto t_end = std::chrono::high_resolution_clock::now();
+
+            printf("stream: whisper_full() returned %d in %f seconds\n", ret, std::chrono::duration<double>(t_end - t_start).count());
+        }
+
+        {
+            std::string text_heard;
+
+            {
+                const int n_segments = whisper_full_n_segments(ctx);
+                for (int i = n_segments - 1; i < n_segments; ++i) {
+                    const char * text = whisper_full_get_segment_text(ctx, i);
+
+                    const int64_t t0 = whisper_full_get_segment_t0(ctx, i);
+                    const int64_t t1 = whisper_full_get_segment_t1(ctx, i);
+
+                    printf("transcribed: %s\n", text);
+
+                    text_heard += text;
+                }
+            }
+
+            {
+                std::lock_guard<std::mutex> lock(g_mutex);
+                g_transcribed = text_heard;
+            }
+        }
+    }
+
+    if (index < g_contexts.size()) {
+        whisper_free(g_contexts[index]);
+        g_contexts[index] = nullptr;
+    }
+}
+
+EMSCRIPTEN_BINDINGS(stream) {
+    emscripten::function("init", emscripten::optional_override([](const std::string & path_model) {
+        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) {
+                    g_running = true;
+                    if (g_worker.joinable()) {
+                        g_worker.join();
+                    }
+                    g_worker = std::thread([i]() {
+                        stream_main(i);
+                    });
+
+                    return i + 1;
+                } else {
+                    return (size_t) 0;
+                }
+            }
+        }
+
+        return (size_t) 0;
+    }));
+
+    emscripten::function("free", emscripten::optional_override([](size_t index) {
+        if (g_running) {
+            g_running = false;
+        }
+    }));
+
+    emscripten::function("set_audio", emscripten::optional_override([](size_t index, const emscripten::val & audio) {
+        --index;
+
+        if (index >= g_contexts.size()) {
+            return -1;
+        }
+
+        if (g_contexts[index] == nullptr) {
+            return -2;
+        }
+
+        {
+            std::lock_guard<std::mutex> lock(g_mutex);
+            const int n = audio["length"].as<int>();
+
+            emscripten::val heap = emscripten::val::module_property("HEAPU8");
+            emscripten::val memory = heap["buffer"];
+
+            g_pcmf32.resize(n);
+
+            emscripten::val memoryView = audio["constructor"].new_(memory, reinterpret_cast<uintptr_t>(g_pcmf32.data()), n);
+            memoryView.call<void>("set", audio);
+        }
+
+        return 0;
+    }));
+
+    emscripten::function("get_transcribed", emscripten::optional_override([]() {
+        std::string transcribed;
+
+        {
+            std::lock_guard<std::mutex> lock(g_mutex);
+            transcribed = std::move(g_transcribed);
+        }
+
+        return transcribed;
+    }));
+
+    emscripten::function("get_status", emscripten::optional_override([]() {
+        std::string status;
+
+        {
+            std::lock_guard<std::mutex> lock(g_mutex);
+            status = g_status_forced.empty() ? g_status : g_status_forced;
+        }
+
+        return status;
+    }));
+
+    emscripten::function("set_status", emscripten::optional_override([](const std::string & status) {
+        {
+            std::lock_guard<std::mutex> lock(g_mutex);
+            g_status_forced = status;
+        }
+    }));
+}

examples/stream.wasm/index-tmpl.html

@@ -0,0 +1,385 @@
+<!doctype html>
+<html lang="en-us">
+    <head>
+        <title>stream : Real-time Whisper transcription in WebAssembly</title>
+
+        <style>
+            #output {
+                width: 100%;
+                height: 100%;
+                margin: 0 auto;
+                margin-top: 10px;
+                border-left: 0px;
+                border-right: 0px;
+                padding-left: 0px;
+                padding-right: 0px;
+                display: block;
+                background-color: black;
+                color: white;
+                font-size: 10px;
+                font-family: 'Lucida Console', Monaco, monospace;
+                outline: none;
+                white-space: pre;
+                overflow-wrap: normal;
+                overflow-x: scroll;
+            }
+        </style>
+    </head>
+    <body>
+        <div id="main-container">
+            <b>stream : Real-time Whisper transcription in WebAssembly</b>
+
+            <br><br>
+
+            You can find more about this project on <a href="https://github.com/ggerganov/whisper.cpp/tree/master/examples/stream.wasm">GitHub</a>.
+
+            <br><br>
+
+            <hr>
+
+            Select the model you would like to use, click the "Start" button and start speaking
+
+            <br><br>
+
+            <div id="model-whisper">
+                Whisper model: <span id="model-whisper-status"></span>
+                <button id="fetch-whisper-tiny-en" onclick="loadWhisper('tiny.en')">tiny.en (75 MB)</button>
+                <button id="fetch-whisper-base-en" onclick="loadWhisper('base.en')">base.en (142 MB)</button>
+                <span id="fetch-whisper-progress"></span>
+
+                <!--
+                    <input type="file" id="file" name="file" onchange="loadFile(event, 'whisper.bin')" />
+                -->
+            </div>
+
+            <br>
+
+            <div id="input">
+                <button id="start"  onclick="onStart()" disabled>Start</button>
+                <button id="stop"   onclick="onStop()" disabled>Stop</button>
+                <button id="clear"  onclick="clearCache()">Clear Cache</button>
+            </div>
+
+            <br>
+
+            <div id="state">
+                Status: <b><span id="state-status">not started</span></b>
+
+                <pre id="state-transcribed">[The transcribed text will be displayed here]</pre>
+            </div>
+
+            <hr>
+
+            Debug output:
+            <textarea id="output" rows="20"></textarea>
+
+            <br>
+
+            <b>Troubleshooting</b>
+
+            <br><br>
+
+            The page does some heavy computations, so make sure:
+
+            <ul>
+                <li>To use a modern web browser (e.g. Chrome, Firefox)</li>
+                <li>To use a fast desktop or laptop computer (i.e. not a mobile phone)</li>
+                <li>Your browser supports WASM <a href="https://webassembly.org/roadmap/">Fixed-width SIMD</a></li>
+            </ul>
+
+            <div class="cell-version">
+                <span>
+                    |
+                    Build time: <span class="nav-link">@GIT_DATE@</span> |
+                    Commit hash: <a class="nav-link" href="https://github.com/ggerganov/whisper.cpp/commit/@GIT_SHA1@">@GIT_SHA1@</a> |
+                    Commit subject: <span class="nav-link">@GIT_COMMIT_SUBJECT@</span> |
+                    <a class="nav-link" href="https://github.com/ggerganov/whisper.cpp/tree/master/examples/stream.wasm">Source Code</a> |
+                </span>
+            </div>
+        </div>
+
+        <script type="text/javascript" src="helpers.js"></script>
+        <script type='text/javascript'>
+            const kRestartRecording_s = 15;
+            const kSampleRate = 16000;
+
+            window.AudioContext = window.AudioContext || window.webkitAudioContext;
+            window.OfflineAudioContext = window.OfflineAudioContext || window.webkitOfflineAudioContext;
+
+            // web audio context
+            var context = null;
+
+            // audio data
+            var audio = null;
+            var audio0 = null;
+
+            // the stream instance
+            var instance = null;
+
+            // model name
+            var model_whisper = null;
+
+            var Module = {
+                print: printTextarea,
+                printErr: printTextarea,
+                setStatus: function(text) {
+                    printTextarea('js: ' + text);
+                },
+                monitorRunDependencies: function(left) {
+                },
+                preRun: function() {
+                    printTextarea('js: Preparing ...');
+                },
+                postRun: function() {
+                    printTextarea('js: Initialized successfully!');
+                }
+            };
+
+            //
+            // fetch models
+            //
+
+            let dbVersion = 1
+            let dbName    = 'whisper.ggerganov.com';
+            let indexedDB = window.indexedDB || window.mozIndexedDB || window.webkitIndexedDB || window.msIndexedDB
+
+            function storeFS(fname, buf) {
+                // write to WASM file using FS_createDataFile
+                // if the file exists, delete it
+                try {
+                    Module.FS_unlink(fname);
+                } catch (e) {
+                    // ignore
+                }
+
+                Module.FS_createDataFile("/", fname, buf, true, true);
+
+                printTextarea('storeFS: stored model: ' + fname + ' size: ' + buf.length);
+
+                document.getElementById('model-whisper-status').innerHTML = 'loaded "' + model_whisper + '"!';
+
+                if (model_whisper != null) {
+                    document.getElementById('start').disabled = false;
+                    document.getElementById('stop' ).disabled = true;
+                }
+            }
+
+            function loadWhisper(model) {
+                let urls = {
+                    'tiny.en': 'https://whisper.ggerganov.com/ggml-model-whisper-tiny.en.bin',
+                    'base.en': 'https://whisper.ggerganov.com/ggml-model-whisper-base.en.bin',
+                };
+
+                let sizes = {
+                    'tiny.en': 75,
+                    'base.en': 142,
+                };
+
+                let url     = urls[model];
+                let dst     = 'whisper.bin';
+                let size_mb = sizes[model];
+
+                model_whisper = model;
+
+                document.getElementById('fetch-whisper-tiny-en').style.display = 'none';
+                document.getElementById('fetch-whisper-base-en').style.display = 'none';
+                document.getElementById('model-whisper-status').innerHTML = 'loading "' + model + '" ... ';
+
+                cbProgress = function(p) {
+                    let el = document.getElementById('fetch-whisper-progress');
+                    el.innerHTML = Math.round(100*p) + '%';
+                };
+
+                cbCancel = function() {
+                    var el;
+                    el = document.getElementById('fetch-whisper-tiny-en'); if (el) el.style.display = 'inline-block';
+                    el = document.getElementById('fetch-whisper-base-en'); if (el) el.style.display = 'inline-block';
+                    el = document.getElementById('model-whisper-status');  if (el) el.innerHTML = '';
+                };
+
+                loadRemote(url, dst, size_mb, cbProgress, storeFS, cbCancel, printTextarea);
+            }
+
+            //
+            // microphone
+            //
+
+            var mediaRecorder = null;
+            var doRecording = false;
+            var startTime = 0;
+
+            function stopRecording() {
+                Module.set_status("paused");
+                doRecording = false;
+                audio0 = null;
+                audio = null;
+                context = null;
+            }
+
+            function startRecording() {
+                if (!context) {
+                    context = new AudioContext({
+                        sampleRate: 16000,
+                        channelCount: 1,
+                        echoCancellation: false,
+                        autoGainControl:  true,
+                        noiseSuppression: true,
+                    });
+                }
+
+                Module.set_status("");
+
+                document.getElementById('start').disabled = true;
+                document.getElementById('stop').disabled = false;
+
+                doRecording = true;
+                startTime = Date.now();
+
+                var chunks = [];
+                var stream = null;
+
+                navigator.mediaDevices.getUserMedia({audio: true, video: false})
+                    .then(function(s) {
+                        stream = s;
+                        mediaRecorder = new MediaRecorder(stream);
+                        mediaRecorder.ondataavailable = function(e) {
+                            chunks.push(e.data);
+
+                            var blob = new Blob(chunks, { 'type' : 'audio/ogg; codecs=opus' });
+                            var reader = new FileReader();
+
+                            reader.onload = function(event) {
+                                var buf = new Uint8Array(reader.result);
+
+                                if (!context) {
+                                    return;
+                                }
+                                context.decodeAudioData(buf.buffer, function(audioBuffer) {
+                                    var offlineContext = new OfflineAudioContext(audioBuffer.numberOfChannels, audioBuffer.length, audioBuffer.sampleRate);
+                                    var source = offlineContext.createBufferSource();
+                                    source.buffer = audioBuffer;
+                                    source.connect(offlineContext.destination);
+                                    source.start(0);
+
+                                    offlineContext.startRendering().then(function(renderedBuffer) {
+                                        audio = renderedBuffer.getChannelData(0);
+
+                                        //printTextarea('js: audio recorded, size: ' + audio.length + ', old size: ' + (audio0 == null ? 0 : audio0.length));
+
+                                        var audioAll = new Float32Array(audio0 == null ? audio.length : audio0.length + audio.length);
+                                        if (audio0 != null) {
+                                            audioAll.set(audio0, 0);
+                                        }
+                                        audioAll.set(audio, audio0 == null ? 0 : audio0.length);
+
+                                        if (instance) {
+                                            Module.set_audio(instance, audioAll);
+                                        }
+                                    });
+                                }, function(e) {
+                                    audio = null;
+                                });
+                            }
+
+                            reader.readAsArrayBuffer(blob);
+                        };
+
+                        mediaRecorder.onstop = function(e) {
+                            if (doRecording) {
+                                setTimeout(function() {
+                                    startRecording();
+                                });
+                            }
+                        };
+
+                        mediaRecorder.start(5000);
+                    })
+                    .catch(function(err) {
+                        printTextarea('js: error getting audio stream: ' + err);
+                    });
+
+                var interval = setInterval(function() {
+                    if (!doRecording) {
+                        clearInterval(interval);
+                        mediaRecorder.stop();
+                        stream.getTracks().forEach(function(track) {
+                            track.stop();
+                        });
+
+                        document.getElementById('start').disabled = false;
+                        document.getElementById('stop').disabled  = true;
+
+                        mediaRecorder = null;
+                    }
+
+                    // if audio length is more than kRestartRecording_s seconds, restart recording
+                    if (audio != null && audio.length > kSampleRate*kRestartRecording_s) {
+                        if (doRecording) {
+                            //printTextarea('js: restarting recording');
+
+                            clearInterval(interval);
+                            audio0 = audio;
+                            audio = null;
+                            mediaRecorder.stop();
+                            stream.getTracks().forEach(function(track) {
+                                track.stop();
+                            });
+                        }
+                    }
+                }, 250);
+            }
+
+            //
+            // main
+            //
+
+            var nLines = 0;
+            var intervalUpdate = null;
+            var transcribedAll = '';
+
+            function onStart() {
+                if (!instance) {
+                    instance = Module.init('whisper.bin');
+
+                    if (instance) {
+                        printTextarea("js: whisper initialized, instance: " + instance);
+                    }
+                }
+
+                if (!instance) {
+                    printTextarea("js: failed to initialize whisper");
+                    return;
+                }
+
+                startRecording();
+
+                intervalUpdate = setInterval(function() {
+                    var transcribed = Module.get_transcribed();
+
+                    if (transcribed != null && transcribed.length > 1) {
+                        transcribedAll += transcribed + '<br>';
+                        nLines++;
+
+                        // if more than 10 lines, remove the first line
+                        if (nLines > 10) {
+                            var i = transcribedAll.indexOf('<br>');
+                            if (i > 0) {
+                                transcribedAll = transcribedAll.substring(i + 4);
+                                nLines--;
+                            }
+                        }
+                    }
+
+                    document.getElementById('state-status').innerHTML = Module.get_status();
+                    document.getElementById('state-transcribed').innerHTML = transcribedAll;
+                }, 100);
+            }
+
+            function onStop() {
+                stopRecording();
+            }
+
+        </script>
+        <script type="text/javascript" src="stream.js"></script>
+    </body>
+</html>

examples/stream/README.md

@@ -21,3 +21,7 @@ brew install sdl2
 
 make stream
 ```
+
+## Web version
+
+This tool can also run in the browser: [examples/stream.wasm](/examples/stream.wasm)

examples/talk.wasm/emscripten.cpp

@@ -61,10 +61,10 @@ void talk_main(size_t index) {
     wparams.print_timestamps = true;
     wparams.print_special    = false;
 
-    wparams.max_tokens           = 32;
+    wparams.max_tokens       = 32;
-    wparams.audio_ctx            = 768; // partial encoder context for better performance
+    wparams.audio_ctx        = 768; // partial encoder context for better performance
 
-    wparams.language             = "en";
+    wparams.language         = "en";
 
     g_gpt2 = gpt2_init("gpt-2.bin");
 

examples/talk.wasm/index-tmpl.html

@@ -504,12 +504,6 @@
                 Module.force_speak(instance);
             }
 
-            async function clearCache() {
-                if (confirm('Are you sure you want to clear the cache?\nAll the models will be downloaded again.')) {
-                    indexedDB.deleteDatabase(dbName);
-                }
-            }
-
             //
             // main
             //