@mux/ai 0.27.1 → 0.28.1

package/dist/workflows/index.d.ts

@@ -1,4 +1,4 @@
-export { A as AskQuestionsOptions, j as AskQuestionsResult, k as AskQuestionsType, l as AudioTranslationOptions, m as AudioTranslationResult, n as AutoCensorProfanityOptions, B as BurnedInCaptionsAnalysis, o as BurnedInCaptionsOptions, p as BurnedInCaptionsPromptOverrides, q as BurnedInCaptionsPromptSections, r as BurnedInCaptionsResult, C as CaptionReplacement, s as CensorMode, u as Chapter, v as ChapterSystemPromptSections, w as ChaptersOptions, x as ChaptersPromptOverrides, y as ChaptersPromptSections, z as ChaptersResult, D as ChaptersType, E as DEFAULT_DESCRIPTION_LENGTH, F as DEFAULT_SUMMARY_KEYWORD_LIMIT, G as DEFAULT_TITLE_LENGTH, H as EditCaptionsOptions, J as EditCaptionsResult, K as EmbeddingsOptions, M as EmbeddingsResult, N as EngagementInsightsOptions, O as EngagementInsightsResult, P as GOOGLE_VISION_LIKELIHOOD_TO_SCORE, Q as HIVE_SEXUAL_CATEGORIES, R as HIVE_VIOLENCE_CATEGORIES, T as HiveModerationOutput, U as HiveModerationSource, V as ModerationOptions, W as ModerationProvider, X as ModerationResult, Y as MomentInsight, Z as OverallInsight, _ as ProfanityDetectionPayload, $ as Question, a0 as QuestionAnswer, a1 as QuestionAnswerType, a2 as ReplacementRecord, a3 as SummarizationOptions, a4 as SummarizationPromptOverrides, a5 as SummarizationPromptSections, a6 as SummaryAndTagsResult, a7 as SummaryType, a8 as ThumbnailModerationScore, a9 as TranslationChunkingOptions, aa as TranslationOptions, ab as TranslationPayload, ac as TranslationResult, ad as aggregateTokenUsage, ae as applyOverrideLists, af as applyReplacements, ag as askQuestions, ah as buildReplacementRegex, ai as burnedInCaptionsSchema, aj as censorVttContent, ak as chapterSchema, al as chaptersSchema, am as createReplacer, an as editCaptions, ao as extractTranscriptSegmentsForHotspots, ap as generateChapters, aq as generateEmbeddings, ar as generateEngagementInsights, as as generateVideoEmbeddings, at as getModerationScores, au as getSummaryAndTags, av as hasBurnedInCaptions, aw as mapShotsToHotspots, ax as normalizeTranslatedVtt, ay as profanityDetectionSchema, az as questionAnswerSchema, aA as shouldSplitChunkTranslationError, aB as summarySchema, aC as transformCueText, aD as translateAudio, aE as translateCaptions, aF as translationSchema } from '../index-Zwh0Bwq3.js';
+export { A as AskQuestionsOptions, j as AskQuestionsResult, k as AskQuestionsType, l as AudioTranslationOptions, m as AudioTranslationResult, n as AutoCensorProfanityOptions, B as BurnedInCaptionsAnalysis, o as BurnedInCaptionsOptions, p as BurnedInCaptionsPromptOverrides, q as BurnedInCaptionsPromptSections, r as BurnedInCaptionsResult, C as CaptionReplacement, s as CensorMode, u as Chapter, v as ChapterSystemPromptSections, w as ChaptersOptions, x as ChaptersPromptOverrides, y as ChaptersPromptSections, z as ChaptersResult, D as ChaptersType, E as DEFAULT_DESCRIPTION_LENGTH, F as DEFAULT_SUMMARY_KEYWORD_LIMIT, G as DEFAULT_TITLE_LENGTH, H as EditCaptionsOptions, J as EditCaptionsResult, K as EmbeddingsOptions, M as EmbeddingsResult, N as EngagementInsightsOptions, O as EngagementInsightsResult, P as GOOGLE_VISION_LIKELIHOOD_TO_SCORE, Q as HIVE_SEXUAL_CATEGORIES, R as HIVE_VIOLENCE_CATEGORIES, T as HiveModerationOutput, U as HiveModerationSource, V as ModerationOptions, W as ModerationProvider, X as ModerationResult, Y as MomentInsight, Z as OverallInsight, _ as ProfanityDetectionPayload, $ as Question, a0 as QuestionAnswer, a1 as QuestionAnswerType, a2 as ReplacementRecord, a3 as SummarizationOptions, a4 as SummarizationPromptOverrides, a5 as SummarizationPromptSections, a6 as SummaryAndTagsResult, a7 as SummaryType, a8 as ThumbnailModerationScore, a9 as TranslationChunkingOptions, aa as TranslationOptions, ab as TranslationPayload, ac as TranslationResult, ad as aggregateTokenUsage, ae as applyOverrideLists, af as applyReplacements, ag as askQuestions, ah as buildReplacementRegex, ai as burnedInCaptionsSchema, aj as censorVttContent, ak as chapterSchema, al as chaptersSchema, am as createReplacer, an as editCaptions, ao as extractTranscriptSegmentsForHotspots, ap as generateChapters, aq as generateEmbeddings, ar as generateEngagementInsights, as as generateVideoEmbeddings, at as getModerationScores, au as getSummaryAndTags, av as hasBurnedInCaptions, aw as mapShotsToHotspots, ax as normalizeTranslatedVtt, ay as profanityDetectionSchema, az as questionAnswerSchema, aA as shouldSplitChunkTranslationError, aB as summarySchema, aC as transformCueText, aD as translateAudio, aE as translateCaptions, aF as translationSchema } from '../index-DUVBpnj5.js';
 import 'zod';
 import '@ai-sdk/anthropic';
 import '@ai-sdk/google';

package/dist/workflows/index.js

@@ -6267,6 +6267,8 @@ async function getSummaryAndTagsInternal(assetId, options, collectedUsage) {
 import { sleep } from "workflow";
 var STATIC_RENDITION_POLL_INTERVAL_MS = 5e3;
 var STATIC_RENDITION_MAX_ATTEMPTS = 36;
+var DUBBING_POLL_INTERVAL_MS = 1e4;
+var DEFAULT_DUBBING_POLL_TIMEOUT_SECONDS = 7200;
 function getReadyAudioStaticRendition(asset) {
   const files = asset.static_renditions?.files;
   if (!files || files.length === 0) {
@@ -6405,9 +6407,16 @@ async function checkElevenLabsDubbingStatus({
     targetLanguages: statusData.target_languages ?? []
   };
 }
-async function downloadDubbedAudioFromElevenLabs({
+async function downloadAndUploadDubbedAudio({
   dubbingId,
   languageCode,
+  assetId,
+  toLanguageCode,
+  s3Endpoint,
+  s3Region,
+  s3Bucket,
+  storageAdapter,
+  s3SignedUrlExpirySeconds,
   credentials
 }) {
   "use step";
@@ -6421,19 +6430,7 @@ async function downloadDubbedAudioFromElevenLabs({
   if (!audioResponse.ok) {
     throw new Error(`Failed to fetch dubbed audio: ${audioResponse.statusText}`);
   }
-  return audioResponse.arrayBuffer();
-}
-async function uploadDubbedAudioToS3({
-  dubbedAudioBuffer,
-  assetId,
-  toLanguageCode,
-  s3Endpoint,
-  s3Region,
-  s3Bucket,
-  storageAdapter,
-  s3SignedUrlExpirySeconds
-}) {
-  "use step";
+  const dubbedAudio = new Uint8Array(await audioResponse.arrayBuffer());
   const s3AccessKeyId = env_default.S3_ACCESS_KEY_ID;
   const s3SecretAccessKey = env_default.S3_SECRET_ACCESS_KEY;
   const audioKey = `audio-translations/${assetId}/auto-to-${toLanguageCode}-${Date.now()}.m4a`;
@@ -6444,7 +6441,7 @@ async function uploadDubbedAudioToS3({
     region: s3Region,
     bucket: s3Bucket,
     key: audioKey,
-    body: new Uint8Array(dubbedAudioBuffer),
+    body: dubbedAudio,
     contentType: "audio/mp4"
   }, storageAdapter);
   const presignedUrl = await createPresignedGetUrlWithStorageAdapter({
@@ -6549,12 +6546,13 @@ async function translateAudio(assetId, toLanguageCode, options = {}) {
     wrapError(error, "Failed to create ElevenLabs dubbing job");
   }
   console.warn("\u23F3 Waiting for dubbing to complete...");
+  const dubbingPollTimeoutSeconds = options.dubbingPollTimeoutSeconds ?? DEFAULT_DUBBING_POLL_TIMEOUT_SECONDS;
+  const maxPollAttempts = Math.max(1, Math.ceil(dubbingPollTimeoutSeconds * 1e3 / DUBBING_POLL_INTERVAL_MS));
   let dubbingStatus = "dubbing";
   let pollAttempts = 0;
-  const maxPollAttempts = 180;
   let targetLanguages = [];
   while (dubbingStatus !== "dubbed" && pollAttempts < maxPollAttempts) {
-    await sleep(1e4);
+    await sleep(DUBBING_POLL_INTERVAL_MS);
     pollAttempts++;
     try {
       const statusResult = await checkElevenLabsDubbingStatus({
@@ -6577,46 +6575,24 @@ async function translateAudio(assetId, toLanguageCode, options = {}) {
   let presignedUrl;
   let uploadedTrackId;
   if (uploadToS3) {
-    console.warn("\u{1F4E5} Downloading dubbed audio from ElevenLabs...");
-    let dubbedAudioBuffer;
+    console.warn("\u{1F4E5} Downloading dubbed audio from ElevenLabs and staging to S3...");
     try {
-      const requestedLangCode = toISO639_3(toLanguageCode);
-      let downloadLangCode = targetLanguages.find(
-        (lang) => lang === requestedLangCode
-      ) ?? targetLanguages.find(
-        (lang) => lang.toLowerCase() === requestedLangCode.toLowerCase()
-      );
-      if (!downloadLangCode && targetLanguages.length > 0) {
-        downloadLangCode = targetLanguages[0];
-        console.warn(`\u26A0\uFE0F Requested language "${requestedLangCode}" not found in target_languages. Using "${downloadLangCode}" instead.`);
-      }
-      if (!downloadLangCode) {
-        downloadLangCode = requestedLangCode;
-        console.warn(`\u26A0\uFE0F No target_languages available from ElevenLabs status. Using requested language code: ${requestedLangCode}`);
-      }
-      dubbedAudioBuffer = await downloadDubbedAudioFromElevenLabs({
+      const downloadLangCode = targetLanguages[0] ?? toISO639_1(toLanguageCode);
+      presignedUrl = await downloadAndUploadDubbedAudio({
         dubbingId,
         languageCode: downloadLangCode,
-        credentials
-      });
-      console.warn("\u2705 Dubbed audio downloaded successfully!");
-    } catch (error) {
-      wrapError(error, "Failed to download dubbed audio");
-    }
-    console.warn("\u{1F4E4} Uploading dubbed audio to S3-compatible storage...");
-    try {
-      presignedUrl = await uploadDubbedAudioToS3({
-        dubbedAudioBuffer,
         assetId,
         toLanguageCode,
         s3Endpoint,
         s3Region,
         s3Bucket,
         storageAdapter: effectiveStorageAdapter,
-        s3SignedUrlExpirySeconds: options.s3SignedUrlExpirySeconds
+        s3SignedUrlExpirySeconds: options.s3SignedUrlExpirySeconds,
+        credentials
       });
+      console.warn("\u2705 Dubbed audio staged to S3 successfully!");
     } catch (error) {
-      wrapError(error, "Failed to upload audio to S3");
+      wrapError(error, "Failed to download and upload dubbed audio");
     }
     if (uploadToMux) {
       console.warn("\u{1F4F9} Adding dubbed audio track to Mux asset...");