Piping MIC recordings from a WebBrowser to C# Form

An example of an HTML page that sends recording remotely to a c# processor.

The file is set for 16khz ready for LLM translation. This takes a file from an HTTP post and processes it in memory (you can uncomment to save copies of the file if needed) and passes it to the LLM function.

<!DOCTYPE html>
<html>
<head>
    <title>Voice Recorder (16kHz)</title>
</head>
<body>
    <h2>Record & Send Audio (16kHz)</h2>
    <button id="startBtn">Start Recording</button>
    <button id="stopBtn" disabled>Stop Recording</button>
    <script>
        let mediaStream;
        let mediaStreamSource;
        let audioContext;
        let scriptProcessor;
        let audioBuffer = [];

        document.getElementById('startBtn').onclick = async () => {
            audioBuffer = [];

            mediaStream = await navigator.mediaDevices.getUserMedia({ audio: true });
            audioContext = new AudioContext({ sampleRate: 44100 });

            mediaStreamSource = audioContext.createMediaStreamSource(mediaStream);

            scriptProcessor = audioContext.createScriptProcessor(4096, 1, 1);
            mediaStreamSource.connect(scriptProcessor);
            scriptProcessor.connect(audioContext.destination);

            scriptProcessor.onaudioprocess = e => {
                const input = e.inputBuffer.getChannelData(0);
                audioBuffer.push(new Float32Array(input));
            };

            document.getElementById('startBtn').disabled = true;
            document.getElementById('stopBtn').disabled = false;
        };

        document.getElementById('stopBtn').onclick = async () => {
            scriptProcessor.disconnect();
            mediaStreamSource.disconnect();
            mediaStream.getTracks().forEach(track => track.stop());

            const mergedBuffer = flattenBuffers(audioBuffer);
            const downsampled = await resampleBuffer(mergedBuffer, audioContext.sampleRate, 16000);
            const wavBlob = encodeWAV(downsampled, 16000);

            const formData = new FormData();
            formData.append('audio', wavBlob, 'recording.wav');

            await fetch('http://{remoteIP}:5001/api/audio/upload', {
                method: 'POST',
                body: formData
            });

            console.log("Audio sent to server.");
            document.getElementById('startBtn').disabled = false;
            document.getElementById('stopBtn').disabled = true;
        };

        function flattenBuffers(buffers) {
            const length = buffers.reduce((acc, curr) => acc + curr.length, 0);
            const result = new Float32Array(length);
            let offset = 0;
            for (let b of buffers) {
                result.set(b, offset);
                offset += b.length;
            }
            return result;
        }

        async function resampleBuffer(buffer, originalRate, targetRate) {
            const offlineCtx = new OfflineAudioContext(1, buffer.length * targetRate / originalRate, targetRate);
            const srcBuffer = offlineCtx.createBuffer(1, buffer.length, originalRate);
            srcBuffer.copyToChannel(buffer, 0);

            const source = offlineCtx.createBufferSource();
            source.buffer = srcBuffer;
            source.connect(offlineCtx.destination);
            source.start(0);

            const rendered = await offlineCtx.startRendering();
            return rendered.getChannelData(0);
        }

        function encodeWAV(samples, sampleRate) {
            const buffer = new ArrayBuffer(44 + samples.length * 2);
            const view = new DataView(buffer);

            function writeString(view, offset, string) {
                for (let i = 0; i < string.length; i++) {
                    view.setUint8(offset + i, string.charCodeAt(i));
                }
            }

            function floatTo16BitPCM(output, offset, input) {
                for (let i = 0; i < input.length; i++) {
                    let s = Math.max(-1, Math.min(1, input[i]));
                    s = s < 0 ? s * 0x8000 : s * 0x7FFF;
                    output.setInt16(offset + i * 2, s, true);
                }
            }

            writeString(view, 0, 'RIFF');
            view.setUint32(4, 36 + samples.length * 2, true);
            writeString(view, 8, 'WAVE');
            writeString(view, 12, 'fmt ');
            view.setUint32(16, 16, true); // PCM chunk size
            view.setUint16(20, 1, true);  // format: PCM
            view.setUint16(22, 1, true);  // mono
            view.setUint32(24, sampleRate, true);
            view.setUint32(28, sampleRate * 2, true); // byte rate
            view.setUint16(32, 2, true); // block align
            view.setUint16(34, 16, true); // bits per sample
            writeString(view, 36, 'data');
            view.setUint32(40, samples.length * 2, true);

            floatTo16BitPCM(view, 44, samples);
            return new Blob([view], { type: 'audio/wav' });
        }
    </script>
</body>
</html>

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

<!DOCTYPE html>

<html>

<head>

<title>Voice Recorder (16kHz)</title>

</head>

<body>

<h2>Record & Send Audio (16kHz)</h2>

<button id="startBtn">Start Recording</button>

<button id="stopBtn" disabled>Stop Recording</button>

let mediaStream;

let mediaStreamSource;

let audioContext;

let scriptProcessor;

let audioBuffer = [];

document.getElementById('startBtn').onclick = async () => {

audioBuffer = [];

mediaStream = await navigator.mediaDevices.getUserMedia({ audio: true });

audioContext = new AudioContext({ sampleRate: 44100 });

mediaStreamSource = audioContext.createMediaStreamSource(mediaStream);

scriptProcessor = audioContext.createScriptProcessor(4096, 1, 1);

mediaStreamSource.connect(scriptProcessor);

scriptProcessor.connect(audioContext.destination);

scriptProcessor.onaudioprocess = e => {

const input = e.inputBuffer.getChannelData(0);

audioBuffer.push(new Float32Array(input));

};

document.getElementById('startBtn').disabled = true;

document.getElementById('stopBtn').disabled = false;

};

document.getElementById('stopBtn').onclick = async () => {

scriptProcessor.disconnect();

mediaStreamSource.disconnect();

mediaStream.getTracks().forEach(track => track.stop());

const mergedBuffer = flattenBuffers(audioBuffer);

const downsampled = await resampleBuffer(mergedBuffer, audioContext.sampleRate, 16000);

const wavBlob = encodeWAV(downsampled, 16000);

const formData = new FormData();

formData.append('audio', wavBlob, 'recording.wav');

await fetch('http://{remoteIP}:5001/api/audio/upload', {

method: 'POST',

body: formData

});

console.log("Audio sent to server.");

document.getElementById('startBtn').disabled = false;

document.getElementById('stopBtn').disabled = true;

};

function flattenBuffers(buffers) {

const length = buffers.reduce((acc, curr) => acc + curr.length, 0);

const result = new Float32Array(length);

let offset = 0;

for (let b of buffers) {

result.set(b, offset);

offset += b.length;

}

return result;

}

async function resampleBuffer(buffer, originalRate, targetRate) {

const offlineCtx = new OfflineAudioContext(1, buffer.length * targetRate / originalRate, targetRate);

const srcBuffer = offlineCtx.createBuffer(1, buffer.length, originalRate);

srcBuffer.copyToChannel(buffer, 0);

const source = offlineCtx.createBufferSource();

source.buffer = srcBuffer;

source.connect(offlineCtx.destination);

source.start(0);

const rendered = await offlineCtx.startRendering();

return rendered.getChannelData(0);

}

function encodeWAV(samples, sampleRate) {

const buffer = new ArrayBuffer(44 + samples.length * 2);

const view = new DataView(buffer);

function writeString(view, offset, string) {

for (let i = 0; i < string.length; i++) {

view.setUint8(offset + i, string.charCodeAt(i));

}

function floatTo16BitPCM(output, offset, input) {

for (let i = 0; i < input.length; i++) {

let s = Math.max(-1, Math.min(1, input[i]));

s = s < 0 ? s * 0x8000 : s * 0x7FFF;

output.setInt16(offset + i * 2, s, true);

}

writeString(view, 0, 'RIFF');

view.setUint32(4, 36 + samples.length * 2, true);

writeString(view, 8, 'WAVE');

writeString(view, 12, 'fmt ');

view.setUint32(16, 16, true); // PCM chunk size

view.setUint16(20, 1, true); // format: PCM

view.setUint16(22, 1, true); // mono

view.setUint32(24, sampleRate, true);

view.setUint32(28, sampleRate * 2, true); // byte rate

view.setUint16(32, 2, true); // block align

view.setUint16(34, 16, true); // bits per sample

writeString(view, 36, 'data');

view.setUint32(40, samples.length * 2, true);

floatTo16BitPCM(view, 44, samples);

return new Blob([view], { type: 'audio/wav' });

}

</script>

</body>

</html>

private WaveInEvent waveIn;
private BufferedWaveProvider buffer;
private WaveFileWriter writer;
private string tempWavFile = Path.Combine(Path.GetTempPath(), "live.wav");
public Form1()
{
    InitializeComponent();
}

private void Form1_Load(object sender, EventArgs e)
{
    StartHttpServer();
    //WAVToText(); //File to Text
    //StartRealtimeIntervalWhisperAsync(); //Every 5 seconds interval
    //StartSilenceAwareWhisperAsync(); //Continus recording
}

private HttpListener listener;
private Thread listenerThread;
private const string Prefix = "http://+:5001/api/audio/upload/";
private void StartHttpServer()
{
    listener = new HttpListener();
    listener.Prefixes.Add(Prefix);
    listener.Start();

    listenerThread = new Thread(() =>
    {
        while (listener.IsListening)
        {
            try
            {
                var context = listener.GetContext();
                ProcessRequest(context);
            }
            catch (HttpListenerException) { break; } // listener closed
            catch (Exception ex)
            {
                Invoke(new Action(() => Log($"Error: {ex.Message}")));
            }
        }
    });
    listenerThread.Start();

    Log("Listening for audio uploads on:");
    Log(Prefix);
}

private int IndexOf(byte[] buffer, byte[] pattern, int startIndex = 0)
{
    for (int i = startIndex; i <= buffer.Length - pattern.Length; i++)
    {
        bool match = true;
        for (int j = 0; j < pattern.Length; j++)
        {
            if (buffer[i + j] != pattern[j])
            {
                match = false;
                break;
            }
        }
        if (match) return i;
    }
    return -1;
}


private async void ProcessRequest(HttpListenerContext context)
{
    if (context.Request.HttpMethod != "POST" ||
        !context.Request.HasEntityBody ||
        !context.Request.ContentType.StartsWith("multipart/form-data"))
    {
        context.Response.StatusCode = 400;
        context.Response.Close();
        return;
    }

    try
    {
        string boundary = "--" + context.Request.ContentType.Split('=')[1];
        byte[] buffer = new BinaryReader(context.Request.InputStream).ReadBytes((int)context.Request.ContentLength64);

        int headerEnd = IndexOf(buffer, Encoding.UTF8.GetBytes("\r\n\r\n")) + 4;
        int boundaryIndex = IndexOf(buffer, Encoding.UTF8.GetBytes(boundary), headerEnd);

        if (headerEnd < 4 || boundaryIndex < 0)
        {
            Log("Could not locate multipart body.");
            context.Response.StatusCode = 400;
            context.Response.Close();
            return;
        }

        int contentLength = boundaryIndex - headerEnd - 2; // remove trailing \r\n
        byte[] fileData = new byte[contentLength];
        Array.Copy(buffer, headerEnd, fileData, 0, contentLength);

        string fileName = $"recording_{DateTime.Now:yyyyMMdd_HHmmss}.wav";
        //string path = Path.Combine(Application.StartupPath, fileName);
        string path = Path.Combine(Path.GetTempPath(), fileName);
        //File.WriteAllBytes(path, fileData);
        context.Response.StatusCode = 200;
        context.Response.AddHeader("Access-Control-Allow-Origin", "*");
        context.Response.Close();

        string ReturnedText = await ByteWAVToText(fileData);
        Log($"Saved: {path}");
        Log($"Saved: {ReturnedText}");
    }
    catch (Exception ex)
    {
        Log("Error: " + ex.Message);
        context.Response.StatusCode = 500;
        context.Response.Close();
    }
}
private void Log(string message)
{
    Debug.WriteLine($"{DateTime.Now:T} - {message}");
}
protected override void OnFormClosing(FormClosingEventArgs e)
{
    listener?.Stop();
    listenerThread?.Join();
    base.OnFormClosing(e);
}

private static async Task<string> ByteWAVToText(byte[] wavBytes)
{
    string StringToReturn = "";
    var ggmlType = GgmlType.Base;
    var modelFileName = "ggml-base.bin";

    IDisposable whisperLogger = null;
    try
    {
        whisperLogger = LogProvider.AddConsoleLogging(WhisperLogLevel.Debug);

        if (!File.Exists(modelFileName))
        {
            await DownloadModel(modelFileName, ggmlType);
        }

        using (var whisperFactory = WhisperFactory.FromPath(modelFileName))
        using (var processor = whisperFactory.CreateBuilder()
            .WithLanguage("auto")
            .Build())
        using (var memoryStream = new MemoryStream(wavBytes))
        {
            var results = processor.ProcessAsync(memoryStream);
            var enumerator = results.GetAsyncEnumerator();

            try
            {
                while (await enumerator.MoveNextAsync())
                {
                    var result = enumerator.Current;
                    Console.WriteLine($"{result.Start}->{result.End}: {result.Text}");
                    StringToReturn += result.Text;
                }
            }
            finally
            {
                await enumerator.DisposeAsync();
            }
        }
    }
    finally
    {
        whisperLogger?.Dispose();
    }
    return StringToReturn;
}

//Validated code for standalone
private static async void WAVFileToText(string wavFileName = @"D:\ffmpeg-7.1.1-full_build\bin\out.wav")
{
    var ggmlType = GgmlType.Base;
    var modelFileName = "ggml-base.bin";

    IDisposable whisperLogger = null;
    try
    {
        whisperLogger = LogProvider.AddConsoleLogging(WhisperLogLevel.Debug);

        if (!File.Exists(modelFileName))
        {
            await DownloadModel(modelFileName, ggmlType);
        }

        using (var whisperFactory = WhisperFactory.FromPath(modelFileName))
        using (var processor = whisperFactory.CreateBuilder()
            .WithLanguage("auto")
            .Build())
        using (var fileStream = File.OpenRead(wavFileName))
        {
            var results = processor.ProcessAsync(fileStream);

            var enumerator = results.GetAsyncEnumerator();
            try
            {
                while (await enumerator.MoveNextAsync())
                {
                    var result = enumerator.Current;
                    Console.WriteLine($"{result.Start}->{result.End}: {result.Text}");
                }
            }
            finally
            {
                if (enumerator != null)
                    await enumerator.DisposeAsync();
            }
        }
    }
    finally
    {
        if (whisperLogger != null)
            whisperLogger.Dispose();
    }
}

private bool _isRecording = false;
private CancellationTokenSource _cancellationSource;

private bool _isSilenceModeActive = false;
private CancellationTokenSource _silenceToken;
private const string modelFileName = "ggml-base.bin";

private async Task StartSilenceAwareWhisperAsync()
{
    var ggmlType = GgmlType.Base;
    var silenceThreshold = 0.005f;
    var silenceDuration = TimeSpan.FromSeconds(1); //After one second of silence, process the request
    IDisposable whisperLogger = null;

    if (!File.Exists(modelFileName))
        await DownloadModel(modelFileName, ggmlType);

    _isSilenceModeActive = true;
    _silenceToken = new CancellationTokenSource();

    try
    {
        whisperLogger = LogProvider.AddConsoleLogging(WhisperLogLevel.None); // Disable verbose logs

        while (_isSilenceModeActive && !_silenceToken.IsCancellationRequested)
        {
            var waveFormat = new WaveFormat(16000, 1);
            var waveIn = new WaveInEvent
            {
                WaveFormat = waveFormat,
                BufferMilliseconds = 200
            };

            var chunkBuffer = new MemoryStream();
            var writer = new WaveFileWriter(chunkBuffer, waveFormat);
            TimeSpan silenceTimer = TimeSpan.Zero;
            bool recordingStopped = false;

            EventHandler<WaveInEventArgs> dataAvailableHandler = (s, a) =>
            {
                writer.Write(a.Buffer, 0, a.BytesRecorded);
                writer.Flush();

                float maxVolume = GetMaxVolume(a.Buffer, a.BytesRecorded);
                if (maxVolume < silenceThreshold)
                {
                    silenceTimer += TimeSpan.FromMilliseconds(waveIn.BufferMilliseconds);
                    if (silenceTimer >= silenceDuration && !recordingStopped)
                    {
                        recordingStopped = true;
                        waveIn.StopRecording();
                    }
                }
                else
                {
                    silenceTimer = TimeSpan.Zero;
                }
            };

            waveIn.DataAvailable += dataAvailableHandler;
            waveIn.StartRecording();

            while (!recordingStopped && !_silenceToken.IsCancellationRequested)
                await Task.Delay(100);

            waveIn.DataAvailable -= dataAvailableHandler;
            writer.Flush();

            byte[] audioData = chunkBuffer.ToArray();
            writer.Dispose();
            waveIn.Dispose();

            if (IsSilent(audioData))
            {
                //Debug.WriteLine("[Skipped silent chunk]");
                continue;
            }

            // Transcribe in background using a dedicated processor
            _ = Task.Run(async () =>
            {
                try
                {
                    using (var whisperFactory = WhisperFactory.FromPath(modelFileName))
                    using (var taskProcessor = whisperFactory.CreateBuilder()
                        .WithLanguage("auto")
                        .Build())
                    using (var memoryStream = new MemoryStream(audioData))
                    {
                        var results = taskProcessor.ProcessAsync(memoryStream);
                        var enumerator = results.GetAsyncEnumerator();
                        try
                        {
                            while (await enumerator.MoveNextAsync())
                            {
                                var result = enumerator.Current;
                                Debug.WriteLine($"{DateTime.Now} {result.Start}->{result.End}:  {result.Text}");
                            }
                        }
                        finally
                        {
                            await enumerator.DisposeAsync();
                        }
                    }
                }
                catch (Exception ex)
                {
                    Debug.WriteLine("Transcription Task Error: " + ex.Message);
                }
            });
        }
    }
    finally
    {
        whisperLogger?.Dispose();
    }
}



private bool IsSilent(byte[] buffer)
{
    if (buffer == null || buffer.Length <= 44)
        return true;

    // Skip the first 44 bytes (WAV header)
    int offset = 44;
    double sumSquares = 0;
    int samples = (buffer.Length - offset) / 2;

    for (int i = offset; i < buffer.Length; i += 2)
    {
        short sample = BitConverter.ToInt16(buffer, i);
        double normalized = sample / 32768.0;
        sumSquares += normalized * normalized;
    }

    double rms = Math.Sqrt(sumSquares / samples);

    // Debug: Show RMS & first few bytes (optional)
    if (false)
    { 
        Debug.WriteLine("RMS: " + rms.ToString("F6"));
        for (int i = 0; i < Math.Min(16, buffer.Length); i++)
            Debug.Write(buffer[i].ToString("X2") + " ");
        Debug.WriteLine("");
    }

    // Silence if RMS is very low
    return rms < 0.001;
}


private float GetMaxVolume(byte[] buffer, int bytesRecorded)
{
    float max = 0;
    for (int i = 0; i < bytesRecorded; i += 2)
    {
        short sample = BitConverter.ToInt16(buffer, i);
        float amplitude = Math.Abs(sample / 32768f);
        if (amplitude > max) max = amplitude;
    }
    return max;
}
private async Task StartRealtimeIntervalWhisperAsync()
{
    var modelFileName = "ggml-base.bin";
    var ggmlType = GgmlType.Base;
    IDisposable whisperLogger = null;

    try
    {
        whisperLogger = LogProvider.AddConsoleLogging(WhisperLogLevel.Warning);

        if (!File.Exists(modelFileName))
        {
            await DownloadModel(modelFileName, ggmlType);
        }

        _isRecording = true;
        _cancellationSource = new CancellationTokenSource();

        using (var whisperFactory = WhisperFactory.FromPath(modelFileName))
        using (var processor = whisperFactory.CreateBuilder()
            .WithLanguage("auto")
            .Build())
        {
            while (_isRecording && !_cancellationSource.IsCancellationRequested)
            {
                string chunkPath = Path.Combine(Path.GetTempPath(), "chunk_" + Guid.NewGuid().ToString("N") + ".wav");

                await RecordChunkAsync(chunkPath, durationMs: 5000);

                using (var fileStream = File.OpenRead(chunkPath))
                {
                    var results = processor.ProcessAsync(fileStream);
                    var enumerator = results.GetAsyncEnumerator();

                    try
                    {
                        while (await enumerator.MoveNextAsync())
                        {
                            var result = enumerator.Current;
                            Debug.WriteLine(result.Start + "->" + result.End + ": " + result.Text);
                        }
                    }
                    finally
                    {
                        if (enumerator != null)
                            await enumerator.DisposeAsync();
                    }
                }

                File.Delete(chunkPath);
            }
        }
    }
    finally
    {
        whisperLogger?.Dispose();
    }
}

private async Task RecordChunkAsync(string filePath, int durationMs)
{
    var waveFormat = new WaveFormat(16000, 1);
    var waveIn = new WaveInEvent
    {
        WaveFormat = waveFormat,
        BufferMilliseconds = 250
    };

    var buffer = new BufferedWaveProvider(waveFormat);
    var writer = new WaveFileWriter(filePath, waveFormat);

    waveIn.DataAvailable += (s, a) =>
    {
        buffer.AddSamples(a.Buffer, 0, a.BytesRecorded);
        writer.Write(a.Buffer, 0, a.BytesRecorded);
        writer.Flush();
    };

    waveIn.StartRecording();
    await Task.Delay(durationMs);
    waveIn.StopRecording();
    writer.Dispose();
    waveIn.Dispose();
}

private void StopRealtimeWhisper()
{
    _isRecording = false;
    _cancellationSource?.Cancel();
}

private static async Task DownloadModel(string fileName, GgmlType ggmlType)
{
    Console.WriteLine($"Downloading Model {fileName}");

    using (var modelStream = await WhisperGgmlDownloader.Default.GetGgmlModelAsync(ggmlType))
    using (var fileWriter = File.OpenWrite(fileName))
    {
        await modelStream.CopyToAsync(fileWriter);
    }
}

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

487

private WaveInEvent waveIn;

private BufferedWaveProvider buffer;

private WaveFileWriter writer;

private string tempWavFile = Path.Combine(Path.GetTempPath(), "live.wav");

public Form1()

{

InitializeComponent();

}

private void Form1_Load(object sender, EventArgs e)

{

StartHttpServer();

//WAVToText(); //File to Text

//StartRealtimeIntervalWhisperAsync(); //Every 5 seconds interval

//StartSilenceAwareWhisperAsync(); //Continus recording

}

private HttpListener listener;

private Thread listenerThread;

private const string Prefix = "http://+:5001/api/audio/upload/";

private void StartHttpServer()

{

listener = new HttpListener();

listener.Prefixes.Add(Prefix);

listener.Start();

listenerThread = new Thread(() =>

{

while (listener.IsListening)

{

try

{

var context = listener.GetContext();

ProcessRequest(context);

}

catch (HttpListenerException) { break; } // listener closed

catch (Exception ex)

{

Invoke(new Action(() => Log($"Error: {ex.Message}")));

}

});

listenerThread.Start();

Log("Listening for audio uploads on:");

Log(Prefix);

}

private int IndexOf(byte[] buffer, byte[] pattern, int startIndex = 0)

{

for (int i = startIndex; i <= buffer.Length - pattern.Length; i++)

{

bool match = true;

for (int j = 0; j < pattern.Length; j++)

{

if (buffer[i + j] != pattern[j])

{

match = false;

break;

}

if (match) return i;

}

return -1;

}

private async void ProcessRequest(HttpListenerContext context)

{

if (context.Request.HttpMethod != "POST" ||

!context.Request.HasEntityBody ||

!context.Request.ContentType.StartsWith("multipart/form-data"))

{

context.Response.StatusCode = 400;

context.Response.Close();

return;

}

try

{

string boundary = "--" + context.Request.ContentType.Split('=')[1];

byte[] buffer = new BinaryReader(context.Request.InputStream).ReadBytes((int)context.Request.ContentLength64);

int headerEnd = IndexOf(buffer, Encoding.UTF8.GetBytes("\r\n\r\n")) + 4;

int boundaryIndex = IndexOf(buffer, Encoding.UTF8.GetBytes(boundary), headerEnd);

if (headerEnd < 4 || boundaryIndex < 0)

{

Log("Could not locate multipart body.");

context.Response.StatusCode = 400;

context.Response.Close();

return;

}

int contentLength = boundaryIndex - headerEnd - 2; // remove trailing \r\n

byte[] fileData = new byte[contentLength];

Array.Copy(buffer, headerEnd, fileData, 0, contentLength);

string fileName = $"recording_{DateTime.Now:yyyyMMdd_HHmmss}.wav";

//string path = Path.Combine(Application.StartupPath, fileName);

string path = Path.Combine(Path.GetTempPath(), fileName);

//File.WriteAllBytes(path, fileData);

context.Response.StatusCode = 200;

context.Response.AddHeader("Access-Control-Allow-Origin", "*");

context.Response.Close();

string ReturnedText = await ByteWAVToText(fileData);

Log($"Saved: {path}");

Log($"Saved: {ReturnedText}");

}

catch (Exception ex)

{

Log("Error: " + ex.Message);

context.Response.StatusCode = 500;

context.Response.Close();

}

private void Log(string message)

{

Debug.WriteLine($"{DateTime.Now:T} - {message}");

}

protected override void OnFormClosing(FormClosingEventArgs e)

{

listener?.Stop();

listenerThread?.Join();

base.OnFormClosing(e);

}

private static async Task<string> ByteWAVToText(byte[] wavBytes)

{

string StringToReturn = "";

var ggmlType = GgmlType.Base;

var modelFileName = "ggml-base.bin";

IDisposable whisperLogger = null;

try

{

whisperLogger = LogProvider.AddConsoleLogging(WhisperLogLevel.Debug);

if (!File.Exists(modelFileName))

{

await DownloadModel(modelFileName, ggmlType);

}

using (var whisperFactory = WhisperFactory.FromPath(modelFileName))

using (var processor = whisperFactory.CreateBuilder()

.WithLanguage("auto")

.Build())

using (var memoryStream = new MemoryStream(wavBytes))

{

var results = processor.ProcessAsync(memoryStream);

var enumerator = results.GetAsyncEnumerator();

try

{

while (await enumerator.MoveNextAsync())

{

var result = enumerator.Current;

Console.WriteLine($"{result.Start}->{result.End}: {result.Text}");

StringToReturn += result.Text;

}

finally

{

await enumerator.DisposeAsync();

}

finally

{

whisperLogger?.Dispose();

}

return StringToReturn;

}

//Validated code for standalone

private static async void WAVFileToText(string wavFileName = @"D:\ffmpeg-7.1.1-full_build\bin\out.wav")

{

var ggmlType = GgmlType.Base;

var modelFileName = "ggml-base.bin";

IDisposable whisperLogger = null;

try

{

whisperLogger = LogProvider.AddConsoleLogging(WhisperLogLevel.Debug);

if (!File.Exists(modelFileName))

{

await DownloadModel(modelFileName, ggmlType);

}

using (var whisperFactory = WhisperFactory.FromPath(modelFileName))

using (var processor = whisperFactory.CreateBuilder()

.WithLanguage("auto")

.Build())

using (var fileStream = File.OpenRead(wavFileName))

{

var results = processor.ProcessAsync(fileStream);

var enumerator = results.GetAsyncEnumerator();

try

{

while (await enumerator.MoveNextAsync())

{

var result = enumerator.Current;

Console.WriteLine($"{result.Start}->{result.End}: {result.Text}");

}

finally

{

if (enumerator != null)

await enumerator.DisposeAsync();

}

finally

{

if (whisperLogger != null)

whisperLogger.Dispose();

}

private bool _isRecording = false;

private CancellationTokenSource _cancellationSource;

private bool _isSilenceModeActive = false;

private CancellationTokenSource _silenceToken;

private const string modelFileName = "ggml-base.bin";

private async Task StartSilenceAwareWhisperAsync()

{

var ggmlType = GgmlType.Base;

var silenceThreshold = 0.005f;

var silenceDuration = TimeSpan.FromSeconds(1); //After one second of silence, process the request

IDisposable whisperLogger = null;

if (!File.Exists(modelFileName))

await DownloadModel(modelFileName, ggmlType);

_isSilenceModeActive = true;

_silenceToken = new CancellationTokenSource();

try

{

whisperLogger = LogProvider.AddConsoleLogging(WhisperLogLevel.None); // Disable verbose logs

while (_isSilenceModeActive && !_silenceToken.IsCancellationRequested)

{

var waveFormat = new WaveFormat(16000, 1);

var waveIn = new WaveInEvent

{

WaveFormat = waveFormat,

BufferMilliseconds = 200

};

var chunkBuffer = new MemoryStream();

var writer = new WaveFileWriter(chunkBuffer, waveFormat);

TimeSpan silenceTimer = TimeSpan.Zero;

bool recordingStopped = false;

EventHandler<WaveInEventArgs> dataAvailableHandler = (s, a) =>

{

writer.Write(a.Buffer, 0, a.BytesRecorded);

writer.Flush();

float maxVolume = GetMaxVolume(a.Buffer, a.BytesRecorded);

if (maxVolume < silenceThreshold)

{

silenceTimer += TimeSpan.FromMilliseconds(waveIn.BufferMilliseconds);

if (silenceTimer >= silenceDuration && !recordingStopped)

{

recordingStopped = true;

waveIn.StopRecording();

}

else

{

silenceTimer = TimeSpan.Zero;

}

};

waveIn.DataAvailable += dataAvailableHandler;

waveIn.StartRecording();

while (!recordingStopped && !_silenceToken.IsCancellationRequested)

await Task.Delay(100);

waveIn.DataAvailable -= dataAvailableHandler;

writer.Flush();

byte[] audioData = chunkBuffer.ToArray();

writer.Dispose();

waveIn.Dispose();

if (IsSilent(audioData))

{

//Debug.WriteLine("[Skipped silent chunk]");

continue;

}

// Transcribe in background using a dedicated processor

_ = Task.Run(async () =>

{

try

{

using (var whisperFactory = WhisperFactory.FromPath(modelFileName))

using (var taskProcessor = whisperFactory.CreateBuilder()

.WithLanguage("auto")

.Build())

using (var memoryStream = new MemoryStream(audioData))

{

var results = taskProcessor.ProcessAsync(memoryStream);

var enumerator = results.GetAsyncEnumerator();

try

{

while (await enumerator.MoveNextAsync())

{

var result = enumerator.Current;

Debug.WriteLine($"{DateTime.Now} {result.Start}->{result.End}: {result.Text}");

}

finally

{

await enumerator.DisposeAsync();

}

catch (Exception ex)

{

Debug.WriteLine("Transcription Task Error: " + ex.Message);

}

});

}

finally

{

whisperLogger?.Dispose();

}

private bool IsSilent(byte[] buffer)

{

if (buffer == null || buffer.Length <= 44)

return true;

// Skip the first 44 bytes (WAV header)

int offset = 44;

double sumSquares = 0;

int samples = (buffer.Length - offset) / 2;

for (int i = offset; i < buffer.Length; i += 2)

{

short sample = BitConverter.ToInt16(buffer, i);

double normalized = sample / 32768.0;

sumSquares += normalized * normalized;

}

double rms = Math.Sqrt(sumSquares / samples);

// Debug: Show RMS & first few bytes (optional)

if (false)

{

Debug.WriteLine("RMS: " + rms.ToString("F6"));

for (int i = 0; i < Math.Min(16, buffer.Length); i++)

Debug.Write(buffer[i].ToString("X2") + " ");

Debug.WriteLine("");

}

// Silence if RMS is very low

return rms < 0.001;

}

private float GetMaxVolume(byte[] buffer, int bytesRecorded)

{

float max = 0;

for (int i = 0; i < bytesRecorded; i += 2)

{

short sample = BitConverter.ToInt16(buffer, i);

float amplitude = Math.Abs(sample / 32768f);

if (amplitude > max) max = amplitude;

}

return max;

}

private async Task StartRealtimeIntervalWhisperAsync()

{

var modelFileName = "ggml-base.bin";

var ggmlType = GgmlType.Base;

IDisposable whisperLogger = null;

try

{

whisperLogger = LogProvider.AddConsoleLogging(WhisperLogLevel.Warning);

if (!File.Exists(modelFileName))

{

await DownloadModel(modelFileName, ggmlType);

}

_isRecording = true;

_cancellationSource = new CancellationTokenSource();

using (var whisperFactory = WhisperFactory.FromPath(modelFileName))

using (var processor = whisperFactory.CreateBuilder()

.WithLanguage("auto")

.Build())

{

while (_isRecording && !_cancellationSource.IsCancellationRequested)

{

string chunkPath = Path.Combine(Path.GetTempPath(), "chunk_" + Guid.NewGuid().ToString("N") + ".wav");

await RecordChunkAsync(chunkPath, durationMs: 5000);

using (var fileStream = File.OpenRead(chunkPath))

{

var results = processor.ProcessAsync(fileStream);

var enumerator = results.GetAsyncEnumerator();

try

{

while (await enumerator.MoveNextAsync())

{

var result = enumerator.Current;

Debug.WriteLine(result.Start + "->" + result.End + ": " + result.Text);

}

finally

{

if (enumerator != null)

await enumerator.DisposeAsync();

}

File.Delete(chunkPath);

}

finally

{

whisperLogger?.Dispose();

}

private async Task RecordChunkAsync(string filePath, int durationMs)

{

var waveFormat = new WaveFormat(16000, 1);

var waveIn = new WaveInEvent

{

WaveFormat = waveFormat,

BufferMilliseconds = 250

};

var buffer = new BufferedWaveProvider(waveFormat);

var writer = new WaveFileWriter(filePath, waveFormat);

waveIn.DataAvailable += (s, a) =>

{

buffer.AddSamples(a.Buffer, 0, a.BytesRecorded);

writer.Write(a.Buffer, 0, a.BytesRecorded);

writer.Flush();

};

waveIn.StartRecording();

await Task.Delay(durationMs);

waveIn.StopRecording();

writer.Dispose();

waveIn.Dispose();

}

private void StopRealtimeWhisper()

{

_isRecording = false;

_cancellationSource?.Cancel();

}

private static async Task DownloadModel(string fileName, GgmlType ggmlType)

{

Console.WriteLine($"Downloading Model {fileName}");

using (var modelStream = await WhisperGgmlDownloader.Default.GetGgmlModelAsync(ggmlType))

using (var fileWriter = File.OpenWrite(fileName))

{

await modelStream.CopyToAsync(fileWriter);

}

Piping MIC recordings from a WebBrowser to C# Form

Leave a comment

Cancel reply