@mux/ai 0.30.0 → 0.31.0

package/dist/{index-D3cB2ltG.d.ts → index-ooikuED7.d.ts}

@@ -1255,6 +1255,10 @@ interface AudioTranslationResult {
     dubbingId: string;
     uploadedTrackId?: string;
     presignedUrl?: string;
+    /** Mux text track ID for the dubbed captions, present when `uploadCaptionsToMux` is true and the upload succeeded. */
+    captionsTrackId?: string;
+    /** Presigned URL for the dub's translated transcript (WebVTT) staged to S3. */
+    captionsPresignedUrl?: string;
     /** Workflow usage metadata (asset duration, thumbnails, etc.). */
     usage?: TokenUsage;
 }
@@ -1290,6 +1294,13 @@ interface AudioTranslationOptions extends MuxAIOptions {
      * required for track creation.
      */
     uploadToMux?: boolean;
+    /**
+     * When true the dub's translated transcript (the same translation that was
+     * voiced) is attached as a subtitles text track on the Mux asset. Implies
+     * `uploadToS3: true`. Defaults to false. The SDK does not check for existing
+     * text tracks — callers decide conflict semantics before enabling this.
+     */
+    uploadCaptionsToMux?: boolean;
     /** Optional storage adapter override for upload + presign operations. */
     storageAdapter?: StorageAdapter;
     /** Expiry duration in seconds for S3 presigned GET URLs. Defaults to 86400 (24 hours). */

package/dist/index.d.ts

@@ -1,4 +1,4 @@
-export { I as ISO639_1, a as ISO639_3, L as LanguageCodePair, S as SupportedISO639_1, b as SupportedISO639_3, g as getLanguageCodePair, c as getLanguageName, i as isUndeterminedLanguageCode, d as isValidISO639_1, e as isValidISO639_3, t as toISO639_1, f as toISO639_3, h as workflows } from './index-D3cB2ltG.js';
+export { I as ISO639_1, a as ISO639_3, L as LanguageCodePair, S as SupportedISO639_1, b as SupportedISO639_3, g as getLanguageCodePair, c as getLanguageName, i as isUndeterminedLanguageCode, d as isValidISO639_1, e as isValidISO639_3, t as toISO639_1, f as toISO639_3, h as workflows } from './index-ooikuED7.js';
 import { W as WorkflowCredentials, S as StoragePutObjectInput, a as StoragePresignGetObjectInput } from './shots-Bh0GrvLx.js';
 export { A as AssetTextTrack, C as ChunkEmbedding, b as ChunkingStrategy, E as Encrypted, c as EncryptedPayload, I as ImageSubmissionMode, M as MuxAIOptions, d as MuxAsset, P as PlaybackAsset, e as PlaybackPolicy, f as ScopedMuxAIOptions, g as StorageAdapter, T as TextChunk, h as TokenChunkingConfig, i as TokenUsage, j as ToneType, U as UsageMetadata, V as VTTChunkingConfig, k as VideoEmbeddingsResult, l as WorkflowCredentialsInput, m as WorkflowMuxClient, n as WorkflowScope, o as decryptFromWorkflow, p as encryptForWorkflow } from './shots-Bh0GrvLx.js';
 import { WORKFLOW_SERIALIZE, WORKFLOW_DESERIALIZE } from '@workflow/serde';
@@ -9,7 +9,7 @@ import '@ai-sdk/google';
 import '@ai-sdk/openai';
 import '@mux/mux-node';
 
-var version = "0.30.0";
+var version = "0.31.0";
 
 type MuxAiErrorType = "validation_error" | "processing_error" | "timeout_error" | "content_policy_error";
 /**

package/dist/index.js

@@ -5,7 +5,7 @@ var __export = (target, all) => {
 };
 
 // package.json
-var version = "0.30.0";
+var version = "0.31.0";
 
 // src/lib/language-codes.ts
 var ISO639_1_TO_3 = {
@@ -3167,13 +3167,23 @@ function createLanguageSection(languageName) {
 }
 
 // src/lib/retry.ts
+import { DownloadError } from "ai";
 var DEFAULT_RETRY_OPTIONS = {
   maxRetries: 3,
   baseDelay: 2e3,
   maxDelay: 1e4
 };
 function defaultShouldRetry(error, _attempt) {
-  return Boolean(error.message && error.message.includes("Timeout while downloading"));
+  if (error.message.includes("Timeout while downloading")) {
+    return true;
+  }
+  const isDownloadError = DownloadError.isInstance(error);
+  const isSerializedDownloadError = error.name === "AI_DownloadError";
+  if (!isDownloadError && !isSerializedDownloadError) {
+    return false;
+  }
+  const statusCode = error.statusCode;
+  return statusCode === void 0 || statusCode === 408 || statusCode === 425 || statusCode === 429 || statusCode >= 500;
 }
 function calculateDelay(attempt, baseDelay, maxDelay) {
   const exponentialDelay = baseDelay * 2 ** (attempt - 1);
@@ -4098,14 +4108,14 @@ async function hasBurnedInCaptionsInternal(assetId, options, collectedUsage) {
       credentials
     });
   } else {
-    analysisResponse = await analyzeStoryboard({
+    analysisResponse = await withRetry(() => analyzeStoryboard({
       imageDataUrl: imageUrl,
       provider: modelConfig.provider,
       modelId: modelConfig.modelId,
       userPrompt,
       systemPrompt: SYSTEM_PROMPT2,
       credentials
-    });
+    }));
   }
   collectedUsage.push(analysisResponse.usage);
   if (!analysisResponse.result) {
@@ -7158,6 +7168,59 @@ async function downloadAndUploadDubbedAudio({
   console.warn(`\u{1F517} Generated presigned URL (expires in ${expiryHours} hour${expiryHours === 1 ? "" : "s"})`);
   return presignedUrl;
 }
+async function downloadAndUploadDubTranscript({
+  dubbingId,
+  languageCode,
+  assetId,
+  toLanguageCode,
+  s3Endpoint,
+  s3Region,
+  s3Bucket,
+  storageAdapter,
+  s3SignedUrlExpirySeconds,
+  credentials
+}) {
+  "use step";
+  const elevenLabsApiKey = await getApiKeyFromEnv("elevenlabs", credentials);
+  const transcriptUrl = `https://api.elevenlabs.io/v1/dubbing/${dubbingId}/transcripts/${languageCode}/format/webvtt`;
+  const transcriptResponse = await fetch(transcriptUrl, {
+    headers: {
+      "xi-api-key": elevenLabsApiKey
+    }
+  });
+  if (!transcriptResponse.ok) {
+    throw new Error(`Failed to fetch dub transcript: ${transcriptResponse.statusText}`);
+  }
+  const transcriptBody = await transcriptResponse.json();
+  const transcriptVtt = transcriptBody.webvtt;
+  if (!transcriptVtt) {
+    throw new Error("Dub transcript response did not include webvtt content");
+  }
+  const s3AccessKeyId = env_default.S3_ACCESS_KEY_ID;
+  const s3SecretAccessKey = env_default.S3_SECRET_ACCESS_KEY;
+  const vttKey = `audio-translations/${assetId}/auto-to-${toLanguageCode}-${Date.now()}.vtt`;
+  await putObjectWithStorageAdapter({
+    accessKeyId: s3AccessKeyId,
+    secretAccessKey: s3SecretAccessKey,
+    endpoint: s3Endpoint,
+    region: s3Region,
+    bucket: s3Bucket,
+    key: vttKey,
+    body: transcriptVtt,
+    contentType: "text/vtt"
+  }, storageAdapter);
+  const presignedUrl = await createPresignedGetUrlWithStorageAdapter({
+    accessKeyId: s3AccessKeyId,
+    secretAccessKey: s3SecretAccessKey,
+    endpoint: s3Endpoint,
+    region: s3Region,
+    bucket: s3Bucket,
+    key: vttKey,
+    expiresInSeconds: s3SignedUrlExpirySeconds ?? 86400
+  }, storageAdapter);
+  console.warn(`\u2705 Dub transcript uploaded successfully to: ${vttKey}`);
+  return presignedUrl;
+}
 async function createAudioTrackOnMux(assetId, languageCode, presignedUrl, credentials) {
   "use step";
   const muxClient = await resolveMuxClient(credentials);
@@ -7184,6 +7247,7 @@ async function translateAudio(assetId, toLanguageCode, options = {}) {
     // 0 = auto-detect
     uploadToS3: uploadToS3Option,
     uploadToMux: uploadToMuxOption,
+    uploadCaptionsToMux = false,
     storageAdapter,
     credentials: providedCredentials
   } = options;
@@ -7193,7 +7257,7 @@ async function translateAudio(assetId, toLanguageCode, options = {}) {
   const credentials = providedCredentials;
   const effectiveStorageAdapter = storageAdapter;
   const uploadToMux = uploadToMuxOption !== false;
-  const uploadToS3 = uploadToS3Option || uploadToMux;
+  const uploadToS3 = uploadToS3Option || uploadToMux || uploadCaptionsToMux;
   const s3Endpoint = options.s3Endpoint ?? env_default.S3_ENDPOINT;
   const s3Region = options.s3Region ?? env_default.S3_REGION ?? "auto";
   const s3Bucket = options.s3Bucket ?? env_default.S3_BUCKET;
@@ -7274,10 +7338,12 @@ async function translateAudio(assetId, toLanguageCode, options = {}) {
   console.warn("\u2705 Dubbing completed successfully!");
   let presignedUrl;
   let uploadedTrackId;
+  let captionsPresignedUrl;
+  let captionsTrackId;
   if (uploadToS3) {
     console.warn("\u{1F4E5} Downloading dubbed audio from ElevenLabs and staging to S3...");
+    const downloadLangCode = targetLanguages[0] ?? toISO639_1(toLanguageCode);
     try {
-      const downloadLangCode = targetLanguages[0] ?? toISO639_1(toLanguageCode);
       presignedUrl = await downloadAndUploadDubbedAudio({
         dubbingId,
         languageCode: downloadLangCode,
@@ -7309,6 +7375,40 @@ async function translateAudio(assetId, toLanguageCode, options = {}) {
         console.warn(presignedUrl);
       }
     }
+    try {
+      console.warn("\u{1F4E5} Downloading dub transcript from ElevenLabs and staging to S3...");
+      captionsPresignedUrl = await downloadAndUploadDubTranscript({
+        dubbingId,
+        languageCode: downloadLangCode,
+        assetId,
+        toLanguageCode,
+        s3Endpoint,
+        s3Region,
+        s3Bucket,
+        storageAdapter: effectiveStorageAdapter,
+        s3SignedUrlExpirySeconds: options.s3SignedUrlExpirySeconds,
+        credentials
+      });
+      if (uploadCaptionsToMux) {
+        console.warn("\u{1F4F9} Adding dubbed captions track to Mux asset...");
+        const muxLangCode = toISO639_1(toLanguageCode);
+        const languageName = new Intl.DisplayNames(["en"], { type: "language" }).of(muxLangCode) || muxLangCode.toUpperCase();
+        captionsTrackId = await createTextTrackOnMux(
+          assetId,
+          muxLangCode,
+          `${languageName} (auto-dubbed)`,
+          captionsPresignedUrl,
+          credentials
+        );
+        console.warn(`\u2705 Captions track added to Mux asset with ID: ${captionsTrackId}`);
+      }
+    } catch (error) {
+      console.warn(`\u26A0\uFE0F Failed to attach dubbed captions: ${error instanceof Error ? error.message : "Unknown error"}`);
+      if (captionsPresignedUrl) {
+        console.warn("\u{1F517} You can manually add the captions track using this presigned URL:");
+        console.warn(captionsPresignedUrl);
+      }
+    }
   }
   const targetLanguage = getLanguageCodePair(toLanguageCode);
   return {
@@ -7318,6 +7418,8 @@ async function translateAudio(assetId, toLanguageCode, options = {}) {
     dubbingId,
     uploadedTrackId,
     presignedUrl,
+    captionsTrackId,
+    captionsPresignedUrl,
     usage: {
       metadata: {
         assetDurationSeconds