feat: implement silence scan for audio segments

README.md

@@ -29,7 +29,9 @@
 - 🔧 **Forced Repair** — Fixes files with corrupted timestamps via an intermediate pipeline (`.w64`/`.mp4`).
 - 🔬 **Quick Scan + Deep Scan** — Checks container integrity and analyzes frame by frame looking for artifacts and errors.
 - 🔬 **Myopic Scan** — Deep Scan restricted to selected tracks.
+- 🔊 **Silence Scan** — Analyze and identify extended periods of silence within audio tracks.
 - 🎛️ **Track Selection** — Choose which video, audio, and subtitle streams to keep in the final file.
+- 🎶 **Smart Spectrum Sync** — Automatically aligns audio tracks from different sources, using advanced waveform cross-correlation.
 - ⏱️ **Sync Adjustment / End Cut** — Defines time offset and end cut to avoid lip-sync issues.
 - 🔀 **File Merging** — Merges tracks from two files into a single MKV, with automatic/manual sync and Strict Mode.
 - 🏷️ **Tag Editing** — Edits language (e.g., `por`, `eng`, `jpn`) and title for each track.
@@ -101,6 +103,7 @@ After analyzing a file, an interactive menu is displayed with the following opti
 - ⏱️ **Adjust Sync / End Cut** — Defines time offset and end cut.
 - 🔍 **Deep Scan** — Frame-by-frame analysis of all tracks.
 - 🔬 **Myopic Scan** — Deep Scan only on selected tracks.
+- 🔊 **Silence Scan** — Detects long periods of silence in audio tracks.
 - 🏷️ **Edit Tags** — Edits language and title of each track.
 
 ## ⚙️ Configuration

README.pt.md

@@ -28,8 +28,10 @@
 - 🔄 **Conversão (Transcode)** — Converte para codecs Direct Play (H.264 8-bit / AAC, EAC3 ou FLAC) com regras de fallback configuráveis.
 - 🔧 **Reparo Forçado** — Corrige arquivos com timestamps corrompidos via pipeline intermediário (`.w64`/`.mp4`).
 - 🔬 **Quick Scan + Deep Scan** — Verifica integridade do container e analisa quadro a quadro em busca de artefatos e erros.
-- 🔬 **Myopic Scan** — Deep Scan restrito às faixas selecionadas.
+- 🔬 **Myopic Scan** — Deep Scan restrito apenas às faixas selecionadas.
+- 🔊 **Silence Scan** — Analisar e identificar longos períodos de silêncio nas faixas de áudio.
 - 🎛️ **Seleção de Faixas** — Escolha quais streams de vídeo, áudio e legenda manter no arquivo final.
+- 🎶 **Smart Spectrum Sync** — Alinha automaticamente faixas de áudio de origens diferentes, usando correlação matemática de ondas sonoras.
 - ⏱️ **Ajuste de Sincronia / Corte Final** — Define offset temporal e corte final para evitar problemas de lip-sync.
 - 🔀 **Mesclagem de Arquivos** — Une faixas de dois arquivos em um único MKV, com sync automático/manual e Modo Estrito.
 - 🏷️ **Edição de Tags** — Edita idioma (ex: `por`, `eng`, `jpn`) e título de cada faixa.
@@ -101,7 +103,8 @@ Após a análise de um arquivo, um menu interativo é exibido com as seguintes o
 - ⏱️ **Ajustar Sincronia / Corte Final** — Define offset temporal e corte final.
 - 🔍 **Deep Scan** — Análise quadro a quadro de todas as faixas.
 - 🔬 **Myopic Scan** — Deep Scan apenas nas faixas selecionadas.
-- 🏷️ **Editar Tags** — Edita idioma e título de cada faixa.
+- 🔊 **Silence Scan** — Detecta longos períodos de silêncio nas faixas de áudio.
+- 🏷️ **Editar Tags** — Edita o idioma e título de cada faixa.
 
 ## ⚙️ Configuração
 

src/commands/check.ts

@@ -13,7 +13,7 @@ import { buildCheckCommand } from '../utils/builder.ts';
 import { ValidationError } from '../utils/errors.ts';
 import { getMediaInfo, runQuickScan } from '../utils/ffprobe.ts';
 import { t } from '../utils/i18n.ts';
-import { filterGarbageStreams } from '../utils/mediaUtils.ts';
+import { filterGarbageStreams, buildAudioMaps, buildAudioLabels, buildSelectedAudioMaps, buildSelectedAudioLabels } from '../utils/mediaUtils.ts';
 import { editTagsMenu, handleExecutionMenu, onCancel, sanitizePath } from '../utils/ui.ts';
 import { renderActionPlan, renderMatrix } from '../views/checkView.ts';
 import { buildGroupedOptions } from '../views/streamOptions.ts';
@@ -35,8 +35,12 @@ interface CheckLoopContext {
   totalFrames: number;
   fullScanInputs: string[];
   fullScanMaps: string[];
+  fullAudioScanMaps: string[];
+  fullAudioScanLabels: string[];
   selectedScanInputs: string[];
   selectedScanMaps: string[];
+  selectedAudioScanMaps: string[];
+  selectedAudioScanLabels: string[];
 }
 
 export async function checkCommand(args: string[]) {
@@ -69,8 +73,12 @@ export async function checkCommand(args: string[]) {
       ffmpegRepairCmd: context.ffmpegRepairCmd,
       fullScanInputs: context.fullScanInputs,
       fullScanMaps: context.fullScanMaps,
+      fullAudioScanMaps: context.fullAudioScanMaps,
+      fullAudioScanLabels: context.fullAudioScanLabels,
       selectedScanInputs: context.selectedScanInputs,
       selectedScanMaps: context.selectedScanMaps,
+      selectedAudioScanMaps: context.selectedAudioScanMaps,
+      selectedAudioScanLabels: context.selectedAudioScanLabels,
       outputPath,
       totalDuration: context.totalDuration,
       totalFrames: context.totalFrames,
@@ -210,8 +218,12 @@ function buildLoopContext(
     totalFrames: diagnostic.metadata.totalFrames,
     fullScanInputs: [targetFile],
     fullScanMaps: ['0'],
+    fullAudioScanMaps: buildAudioMaps(probeData.streams, 0),
+    fullAudioScanLabels: buildAudioLabels(probeData.streams),
     selectedScanInputs: [targetFile],
-    selectedScanMaps: selectedStreams.map((stream) => `0:${stream.streamIndex}`)
+    selectedScanMaps: selectedStreams.map((stream) => `0:${stream.streamIndex}`),
+    selectedAudioScanMaps: buildSelectedAudioMaps(selectedStreams),
+    selectedAudioScanLabels: buildSelectedAudioLabels(selectedStreams)
   };
 }
 

src/commands/merge.ts

@@ -1,6 +1,6 @@
 import fs from 'fs';
 import path from 'path';
-import { confirm, groupMultiselect, log, note, select, text } from '@clack/prompts';
+import { confirm, groupMultiselect, log, note, select, text, spinner } from '@clack/prompts';
 import pc from 'picocolors';
 
 import { getPreferredVideoSource, getPrimaryVideoStream } from '../services/analyzer.ts';
@@ -11,13 +11,18 @@ import type { FFprobeData, SelectedStream } from '../types/media';
 import { buildMergeCommand } from '../utils/builder.ts';
 import { ValidationError } from '../utils/errors.ts';
 import { getMediaInfo } from '../utils/ffprobe.ts';
-import { calculateTotalFrames } from '../utils/formatters.ts';
+import { calculateTotalFrames, parseTimestampToSeconds, formatSecondsToTimestamp } from '../utils/formatters.ts';
 import { t } from '../utils/i18n.ts';
 import { editTagsMenu, handleExecutionMenu, onCancel, sanitizePath } from '../utils/ui.ts';
 import { buildSyncOptions, renderComparison } from '../views/mergeView.ts';
 import { buildGroupedOptions } from '../views/streamOptions.ts';
+import { calculateSpectrumDelay } from '../services/spectrumAnalyzer.ts';
+import { extractRawAudio } from '../utils/ffmpeg.ts';
+import { buildAudioMaps, buildAudioLabels, buildSelectedAudioMaps, buildSelectedAudioLabels } from '../utils/mediaUtils.ts';
 
 const fallbackRules = fallbackRulesData as FallbackRules;
+const SPECTRUM_SAMPLE_WINDOW_SEC = 10;
+const SPECTRUM_SEARCH_WINDOW_SEC = 30;
 
 interface MergeSourcePaths {
   sourcePathA: string;
@@ -54,6 +59,10 @@ interface MergeLoopContext {
   totalFrames: number;
   fullScanInputs: string[];
   fullScanMaps: string[];
+  fullAudioScanMaps: string[];
+  fullAudioScanLabels: string[];
+  selectedAudioScanMaps: string[];
+  selectedAudioScanLabels: string[];
 }
 
 export async function mergeCommand(_args: string[]) {
@@ -72,7 +81,7 @@ export async function mergeCommand(_args: string[]) {
 
   if (Math.abs(media.durationDiffMs) > 1000) {
     note(pc.yellow(t('mergeDurationAlert')), t('durationAlertTitle'));
-    applySyncState(state, await promptSyncAdjustment(media.durationDiffMs, state));
+    applySyncState(state, await promptSyncAdjustment(media, state));
   }
 
   while (true) {
@@ -85,6 +94,10 @@ export async function mergeCommand(_args: string[]) {
       ffmpegRepairCmd: context.ffmpegRepairCmd,
       fullScanInputs: context.fullScanInputs,
       fullScanMaps: context.fullScanMaps,
+      fullAudioScanMaps: context.fullAudioScanMaps,
+      fullAudioScanLabels: context.fullAudioScanLabels,
+      selectedAudioScanMaps: context.selectedAudioScanMaps,
+      selectedAudioScanLabels: context.selectedAudioScanLabels,
       outputPath: media.outputPath,
       totalDuration: context.totalDuration,
       totalFrames: context.totalFrames,
@@ -110,7 +123,7 @@ export async function mergeCommand(_args: string[]) {
     }
 
     if (result.action === 'adjust_sync') {
-      applySyncState(state, await promptSyncAdjustment(media.durationDiffMs, state));
+      applySyncState(state, await promptSyncAdjustment(media, state));
       continue;
     }
 
@@ -223,10 +236,10 @@ function buildStreamOptionSources(media: MergeMediaContext) {
 }
 
 async function promptSyncAdjustment(
-  durationDiffMs: number,
+  media: MergeMediaContext,
   currentState: MergeSyncState
 ): Promise<MergeSyncState> {
-  const options = buildSyncOptions(durationDiffMs);
+  const options = buildSyncOptions(media.durationDiffMs);
 
   const chosenSyncAction = onCancel(await select({
     message: t('mergeHowToSync'),
@@ -236,8 +249,50 @@ async function promptSyncAdjustment(
   let nextDelayMs = currentState.currentDelayMs;
   let nextApplyShortest = currentState.applyShortest;
 
-  if (chosenSyncAction === 'auto') {
-    nextDelayMs = durationDiffMs;
+  if (chosenSyncAction === 'spectrum') {
+    log.info(pc.cyan(t('mergeSpectrumHint')));
+
+    const tsStr = onCancel(await text({
+      message: t('mergeAskTimestamp'),
+      placeholder: '00:15:30',
+      validate(value) {
+        if (!value || !value.includes(':')) return t('validNumber');
+      }
+    }));
+
+    const durA = SPECTRUM_SAMPLE_WINDOW_SEC;
+    const durB = SPECTRUM_SEARCH_WINDOW_SEC; 
+
+    const tsSecs = parseTimestampToSeconds(tsStr);
+    const tsBSecs = Math.max(0, tsSecs - SPECTRUM_SAMPLE_WINDOW_SEC); // Starts before the timestamp
+    const startTsB = formatSecondsToTimestamp(tsBSecs);
+
+    const s = spinner();
+    s.start(t('mergeSpectrumExtracting'));
+
+    try {
+      const bufferA = await extractRawAudio(media.sourcePathA, tsStr, durA);
+      const bufferB = await extractRawAudio(media.sourcePathB, startTsB, durB);
+
+      s.message(t('mergeSpectrumCalculating'));
+
+      const rawOffsetMs = calculateSpectrumDelay(bufferA, bufferB);
+      const diffSecs = tsSecs - tsBSecs;
+
+      const realDelayMs = rawOffsetMs - (diffSecs * 1000);
+      nextDelayMs = Math.round(realDelayMs * -1);
+
+      s.stop(pc.green(t('successOp')));
+      log.success(pc.green(t('mergeSpectrumResult', nextDelayMs)));
+
+    } catch (err) {
+      s.stop(pc.red(t('mergeSpectrumFailed')));
+      if (err instanceof Error) {
+        log.error(err.message);
+      }
+    }
+  } else if (chosenSyncAction === 'auto') {
+    nextDelayMs = media.durationDiffMs;
   } else if (chosenSyncAction === 'manual') {
     const delayStr = onCancel(await text({
       message: t('mergeAskDelay'),
@@ -297,7 +352,11 @@ function buildLoopContext(media: MergeMediaContext, state: MergeSessionState): M
     totalDuration: media.totalDuration,
     totalFrames: media.totalFrames,
     fullScanInputs: [media.sourcePathA, media.sourcePathB],
-    fullScanMaps: buildFullScanMaps(media)
+    fullScanMaps: buildFullScanMaps(media),
+    fullAudioScanMaps: buildFullAudioScanMaps(media),
+    fullAudioScanLabels: buildFullAudioScanLabels(media),
+    selectedAudioScanMaps: buildSelectedAudioMaps(state.selectedStreams),
+    selectedAudioScanLabels: buildSelectedAudioLabels(state.selectedStreams)
   };
 }
 
@@ -308,6 +367,20 @@ function buildFullScanMaps(media: MergeMediaContext) {
   ];
 }
 
+function buildFullAudioScanMaps(media: MergeMediaContext) {
+  return [
+    ...buildAudioMaps(media.infoA.streams, 0),
+    ...buildAudioMaps(media.infoB.streams, 1)
+  ];
+}
+
+function buildFullAudioScanLabels(media: MergeMediaContext) {
+  return [
+    ...buildAudioLabels(media.infoA.streams),
+    ...buildAudioLabels(media.infoB.streams)
+  ];
+}
+
 function buildSourceScanMaps(probeData: FFprobeData, fileIndex: number) {
   return probeData.streams
     .filter((stream) => stream.codec_type === 'video' || stream.codec_type === 'audio')

src/locales/en-US.ts

@@ -17,6 +17,7 @@ export default {
   menuAdjustSync: '⏱️  Adjust Sync / Final Cut',
   menuDeepScanFull: '🔍 Deep Scan (All tracks from original file)',
   menuDeepScanSelected: '🔬 Myopic Scan (Selected tracks only)',
+  menuSilenceScan: '🔊 Silence Scan (Check for audio drops / Mute)',
   menuRunRepairScan: '🔧 Execute Forced Repair + 🔍 Deep Scan',
   menuRunRepairOnly: '🔧 Execute Forced Repair only',
   menuEditTags: '🏷️  Edit Track Tags (Language/Title)',
@@ -79,7 +80,7 @@ export default {
   mergeNone: 'None',
   mergeSelectStreams: 'Select the streams you want to keep',
   mergeHowToSync: 'How do you want to adjust the sync for File B?',
-  mergeAutoSync: '⏭️  Auto-align ({0} to match File A)',
+  mergeAutoSync: '⏭️  Match Durations ({0} offset to match File A)',
   mergeManualSync: '✏️  Enter value manually',
   mergeNoSync: '❌  Do not adjust (0ms)',
   mergeAskDelay: 'Enter the delay for File B in milliseconds (e.g., 2000 to delay 2s, -500 to advance 0.5s):',
@@ -88,6 +89,15 @@ export default {
   mergeModifyStreams: 'Modify the streams you want to keep:',
   mergeCmdSuggested: 'Suggested FFmpeg Command (Merge)',
 
+  // Command: Merge (Spectrum Sync)
+  mergeSpectrumSync: '🎶 Smart Spectrum Sync (Audio-based auto-align)',
+  mergeAskTimestamp: 'Enter the timestamp of File A (Reference) for analysis:',
+  mergeSpectrumHint: '💡 Tip: Choose a timestamp without dialogue, preferably with music or strong ambient sounds.',
+  mergeSpectrumResult: 'Sync complete! Calculated compensation: {0}ms',
+  mergeSpectrumExtracting: 'Extracting audio frequencies for comparison...',
+  mergeSpectrumCalculating: 'Calculating cross-correlation matrix...',
+  mergeSpectrumFailed: 'Failed to analyze audio spectrum.',
+
   // Scanners (FFprobe / FFmpeg)
   scanQuickStart: 'Running integrity Quick Scan...',
   scanQuickPass: '✔ Quick Scan passed: Container structure is intact.',
@@ -107,6 +117,15 @@ export default {
   convFail: '✖ Error during conversion (Code {0}).',
   convStartFail: '✖ Failed to start FFmpeg process: {0}',
 
+  // Silence Scan
+  scanSilenceStart: 'Searching for audio drops (Silence Scan)...',
+  scanSilencePass: '✔ Audio intact: No sudden drops detected.',
+  scanSilenceWarn: '⚠ Warning: Audio drops detected!',
+  scanSilenceDetail: 'Silence detected at {0} (duration: {1}s)',
+  scanSilenceFail: '✖ Failed to execute Silence Scan.',
+  scanSilenceProgress: 'Silence Scan in progress: {0}%',
+  scanSilenceItem: '(duration: {0}s)',
+
   // Command: Lang
   langSelect: '🌐 Select your preferred language:',
   langChanged: '✔ Language changed successfully!',

src/locales/pt-BR.ts

@@ -17,6 +17,7 @@ export default {
   menuAdjustSync: '⏱️  Ajustar Sincronia / Corte Final',
   menuDeepScanFull: '🔍 Deep Scan (Todas as faixas do arquivo original)',
   menuDeepScanSelected: '🔬 Myopic Scan (Apenas faixas selecionadas)',
+  menuSilenceScan: '🔊 Silence Scan (Verificar quedas de áudio / Mudo)',
   menuRunRepairScan: '🔧 Executar Reparo Forçado + 🔍 Deep Scan',
   menuRunRepairOnly: '🔧 Executar Reparo Forçado apenas',
   menuEditTags: '🏷️  Editar Tags das Faixas (Idioma/Título)',
@@ -79,7 +80,7 @@ export default {
   mergeNone: 'Nenhuma',
   mergeSelectStreams: 'Selecione as faixas que deseja manter',
   mergeHowToSync: 'Como deseja ajustar a sincronia do Arquivo B?',
-  mergeAutoSync: '⏭️  Auto-alinhar ({0} para igualar ao Arquivo A)',
+  mergeAutoSync: '⏭️  Igualar Durações ({0} para igualar ao Arquivo A)',
   mergeManualSync: '✏️  Digitar valor manualmente',
   mergeNoSync: '❌  Não ajustar (0ms)',
   mergeAskDelay: 'Informe o atraso do Arquivo B em milissegundos (ex: 2000 para atrasar 2s, -500 para adiantar 0.5s):',
@@ -88,6 +89,15 @@ export default {
   mergeModifyStreams: 'Modifique as faixas que deseja manter:',
   mergeCmdSuggested: 'Comando FFmpeg Sugerido (Merge)',
 
+  // Command: Merge (Spectrum Sync)
+  mergeSpectrumSync: '🎶 Smart Spectrum Sync (Auto-alinhamento por áudio)',
+  mergeAskTimestamp: 'Digite o momento do Arquivo A (Referência) para análise:',
+  mergeSpectrumHint: '💡 Dica: Escolha um momento sem falas/dublagem, de preferência com música ou sons ambientes marcantes.',
+  mergeSpectrumResult: 'Sincronia concluída! Compensação calculada: {0}ms',
+  mergeSpectrumExtracting: 'Extraindo frequências de áudio para comparação...',
+  mergeSpectrumCalculating: 'Calculando matriz de correlação cruzada...',
+  mergeSpectrumFailed: 'Falha ao analisar espectro de áudio.',
+
   // Scanners (FFprobe / FFmpeg)
   scanQuickStart: 'Executando Quick Scan de integridade...',
   scanQuickPass: '✔ Quick Scan aprovado: Estrutura do container intacta.',
@@ -107,6 +117,15 @@ export default {
   convFail: '✖ Erro durante a conversão (Código {0}).',
   convStartFail: '✖ Falha ao tentar iniciar o processo do FFmpeg: {0}',
 
+  // Silence Scan
+  scanSilenceStart: 'Procurando por quedas de áudio (Silence Scan)...',
+  scanSilencePass: '✔ Áudio íntegro: Nenhuma queda brusca detectada.',
+  scanSilenceWarn: '⚠ Atenção: Quedas de áudio detectadas!',
+  scanSilenceDetail: 'Silêncio detectado em {0} (duração: {1}s)',
+  scanSilenceFail: '✖ Falha ao executar o Silence Scan.',
+  scanSilenceProgress: 'Silence Scan em andamento: {0}%',
+  scanSilenceItem: '(duração: {0}s)',
+
   // Comando: Lang
   langSelect: '🌐 Selecione o idioma de sua preferência / Select your preferred language:',
   langChanged: '✔ Idioma alterado com sucesso!',

src/services/spectrumAnalyzer.test.ts

@@ -0,0 +1,40 @@
+import { describe, expect, test } from 'bun:test';
+import { calculateSpectrumDelay } from './spectrumAnalyzer.ts';
+
+describe('services/spectrumAnalyzer.ts', () => {
+  test('Should return 0 for perfect alignment (identical arrays)', () => {
+    const audioA = new Float32Array([0.1, -0.2, 0.5, 0.3]);
+    const audioB = new Float32Array([0.1, -0.2, 0.5, 0.3]);
+
+    expect(calculateSpectrumDelay(audioA, audioB)).toBe(0);
+  });
+
+  test('Should accurately find positive displacement', () => {
+    const audioA = new Float32Array([0.5, 0.8, -0.2]);
+    const audioB = new Float32Array([0, 0, 0.5, 0.8, -0.2, 0.1]);
+
+    expect(calculateSpectrumDelay(audioA, audioB)).toBe(2);
+  });
+
+  test('Should align correctly even with inverted phase (phase shift)', () => {
+    const audioA = new Float32Array([0.5, 0.8, -0.2]);
+    const audioB = new Float32Array([0, 0, -0.5, -0.8, 0.2, 0]);
+
+    expect(calculateSpectrumDelay(audioA, audioB)).toBe(2);
+  });
+
+  test('Should return 0 if snippet has absolute silence (zero variance)', () => {
+    const audioA = new Float32Array([0, 0, 0]);
+    const audioB = new Float32Array([0, 0, 0, 0, 0]);
+
+    expect(calculateSpectrumDelay(audioA, audioB)).toBe(0);
+  });
+
+  test('Should throw error if search window is smaller than snippet', () => {
+    const audioA = new Float32Array([1, 2, 3]);
+    const audioB = new Float32Array([1, 2]);
+
+    expect(() => calculateSpectrumDelay(audioA, audioB))
+      .toThrow('Search window in File B must be larger than File A snippet.');
+  });
+});