Merge pull request #31 from AssemblyAI/E07417BDFEA3614F5967B1520F8B2F61

Release 4.2.1

CHANGELOG.md

@@ -1,5 +1,14 @@
 # Changelog
 
+## [4.2.1] - 2024-01-23
+
+### Added
+- Add `answer_format` to `LemurActionItemsParams` type
+
+### Changed
+- Rename `RealtimeService` to `RealtimeTranscriber`, `RealtimeServiceFactory` to `RealtimeTranscriberFactory`, `RealtimeTranscriberFactory.createService()` to `RealtimeTranscriberFactory.transcriber()`. Deprecated aliases are provided for all old types and functions for backwards compatibility.
+- Restrict the type for `redact_pii_audio_quality` from `string` to `RedactPiiAudioQuality` an enum string.
+
 ## [4.2.0] - 2024-01-11
 
 ### Added

README.md

@@ -184,18 +184,18 @@ const { response } = await client.lemur.task({
 });
 ```
 
-## Transcribe in real time
+## Transcribe in real-time
 
-Create the real-time service.
+Create the real-time transcriber.
 
 ```typescript
-const rt = client.realtime.createService();
+const rt = client.realtime.transcriber();
 ```
 
 You can also pass in the following options.
 
 ```typescript
-const rt = client.realtime.createService({
+const rt = client.realtime.transcriber({
   realtimeUrl: 'wss://localhost/override',
   apiKey: process.env.ASSEMBLYAI_API_KEY // The API key passed to `AssemblyAI` will be used by default,
   sampleRate: 16_000,
@@ -207,7 +207,7 @@ You can also generate a temporary auth token for real-time.
 
 ```typescript
 const token = await client.realtime.createTemporaryToken({ expires_in = 60 });
-const rt = client.realtime.createService({
+const rt = client.realtime.transcriber({
   token: token,
 });
 ```

docs/compat.md

@@ -13,7 +13,7 @@ If you do use an older version of Node.js like version 16, you'll need to polyfi
 To make the SDK compatible with the browser, the SDK aims to use web standards as much as possible.
 However, there are still incompatibilities between Node.js and the browser.
 
-- `RealtimeService` doesn't support the AssemblyAI API key in the browser.
+- `RealtimeTranscriber` doesn't support the AssemblyAI API key in the browser.
   Instead, you have to generate a temporary auth token using `client.realtime.createTemporaryToken`, and pass in the resulting token to the real-time transcriber.
 
   Generate a temporary auth token on the server.
@@ -31,16 +31,16 @@ However, there are still incompatibilities between Node.js and the browser.
   > If you embed the API key on the client, everyone can see it and use it for themselves.
 
   Then pass the token via an API to the client.
-  On the client, create an instance of `RealtimeService` using the token.
+  On the client, create an instance of `RealtimeTranscriber` using the token.
 
   ```js
-  import { RealtimeService } from "assemblyai";
+  import { RealtimeTranscriber } from "assemblyai";
   // or the following if you're using UMD
-  // const { RealtimeService } = assemblyai;
+  // const { RealtimeTranscriber } = assemblyai;
 
   const token = getToken(); // getToken is a function for you to implement
 
-  const rt = new RealtimeService({
+  const rt = new RealtimeTranscriber({
     token: token,
   });
   ```

package.json

@@ -1,6 +1,6 @@
 {
   "name": "assemblyai",
-  "version": "4.2.0",
+  "version": "4.2.1",
   "description": "The AssemblyAI JavaScript SDK provides an easy-to-use interface for interacting with the AssemblyAI API, which supports async and real-time transcription, as well as the latest LeMUR models.",
   "engines": {
     "node": ">=18"

scripts/generate-types.ts

@@ -7,19 +7,8 @@ async function generateTypes(apiSpecPath: string, outputPath: string) {
     alphabetize: true,
     exportType: true,
     transform(schemaObject) {
-      if (
-        "x-fern-type" in schemaObject &&
-        schemaObject["x-fern-type"] === "datetime"
-      ) {
-        // Use Date as type instead of String, even though it will be a string.
-        // The service code manually converts the string into a Date.
-        // Fe see `TranscriptService#list`.
-        return schemaObject.nullable ? "Date | null" : "Date";
-      }
-    },
-    postTransform(type) {
-      if (type === `components["schemas"]["LemurModel"] | string`) {
-        return `LiteralUnion<components["schemas"]["LemurModel"], string>`;
+      if ("x-ts-type" in schemaObject) {
+        return schemaObject["x-ts-type"];
       }
     },
   });

scripts/kitchensink.ts

@@ -7,7 +7,7 @@ import {
   LemurBaseResponse,
   PartialTranscript,
   RealtimeTranscript,
-  CreateRealtimeServiceParams,
+  CreateRealtimeTranscriberParams,
   TranscribeParams,
 } from "../src";
 
@@ -17,7 +17,7 @@ const client = new AssemblyAI({
 
 (async function transcribeUsingRealtime() {
   const useToken = false;
-  const serviceParams: CreateRealtimeServiceParams = {
+  const serviceParams: CreateRealtimeTranscriberParams = {
     sampleRate: 16_000,
     wordBoost: ["gore", "climate"],
     token: useToken
@@ -27,7 +27,7 @@ const client = new AssemblyAI({
       : undefined,
     encoding: "pcm_s16le",
   };
-  const rt = client.realtime.createService(serviceParams);
+  const rt = client.realtime.transcriber(serviceParams);
 
   rt.on("open", ({ sessionId, expiresAt }) => {
     console.log("Session ID:", sessionId, "Expires At:", expiresAt);

src/services/index.ts

@@ -1,6 +1,11 @@
 import { BaseServiceParams } from "..";
 import { LemurService } from "./lemur";
-import { RealtimeService, RealtimeServiceFactory } from "./realtime";
+import {
+  RealtimeTranscriber,
+  RealtimeTranscriberFactory,
+  RealtimeService,
+  RealtimeServiceFactory,
+} from "./realtime";
 import { TranscriptService } from "./transcripts";
 import { FileService } from "./files";
 
@@ -25,7 +30,7 @@ class AssemblyAI {
   /**
    * The realtime service.
    */
-  public realtime: RealtimeServiceFactory;
+  public realtime: RealtimeTranscriberFactory;
 
   /**
    * Create a new AssemblyAI client.
@@ -38,13 +43,15 @@ class AssemblyAI {
     this.files = new FileService(params);
     this.transcripts = new TranscriptService(params, this.files);
     this.lemur = new LemurService(params);
-    this.realtime = new RealtimeServiceFactory(params);
+    this.realtime = new RealtimeTranscriberFactory(params);
   }
 }
 
 export {
   AssemblyAI,
   LemurService,
+  RealtimeTranscriberFactory,
+  RealtimeTranscriber,
   RealtimeServiceFactory,
   RealtimeService,
   TranscriptService,

src/services/realtime/factory.ts

@@ -1,27 +1,35 @@
 import {
   BaseServiceParams,
   RealtimeTokenParams,
-  CreateRealtimeServiceParams,
-  RealtimeServiceParams,
+  CreateRealtimeTranscriberParams,
+  RealtimeTranscriberParams,
   RealtimeTemporaryTokenResponse,
+  CreateRealtimeServiceParams,
 } from "../..";
-import { RealtimeService } from "./service";
+import { RealtimeService, RealtimeTranscriber } from "./service";
 import { BaseService } from "../base";
 
-export class RealtimeServiceFactory extends BaseService {
+export class RealtimeTranscriberFactory extends BaseService {
   private rtFactoryParams: BaseServiceParams;
   constructor(params: BaseServiceParams) {
     super(params);
     this.rtFactoryParams = params;
   }
 
+  /**
+   * @deprecated Use transcriber(...) instead
+   */
   createService(params?: CreateRealtimeServiceParams): RealtimeService {
+    return this.transcriber(params);
+  }
+
+  transcriber(params?: CreateRealtimeTranscriberParams): RealtimeTranscriber {
     const serviceParams = { ...params } as Record<string, unknown>;
     if (!serviceParams.token && !serviceParams.apiKey) {
       serviceParams.apiKey = this.rtFactoryParams.apiKey;
     }
 
-    return new RealtimeService(serviceParams as RealtimeServiceParams);
+    return new RealtimeTranscriber(serviceParams as RealtimeTranscriberParams);
   }
 
   async createTemporaryToken(params: RealtimeTokenParams) {
@@ -35,3 +43,8 @@ export class RealtimeServiceFactory extends BaseService {
     return data.token;
   }
 }
+
+/**
+ * @deprecated Use RealtimeTranscriberFactory instead
+ */
+export class RealtimeServiceFactory extends RealtimeTranscriberFactory {}

src/services/realtime/service.ts

@@ -4,13 +4,14 @@ import { ErrorEvent, MessageEvent, CloseEvent } from "ws";
 import {
   RealtimeEvents,
   RealtimeListeners,
-  RealtimeServiceParams,
+  RealtimeTranscriberParams,
   RealtimeMessage,
   RealtimeTranscript,
   PartialTranscript,
   FinalTranscript,
   SessionBeginsEventData,
   AudioEncoding,
+  AudioData,
 } from "../..";
 import {
   RealtimeError,
@@ -20,7 +21,7 @@ import {
 
 const defaultRealtimeUrl = "wss://api.assemblyai.com/v2/realtime/ws";
 
-export class RealtimeService {
+export class RealtimeTranscriber {
   private realtimeUrl: string;
   private sampleRate: number;
   private wordBoost?: string[];
@@ -31,7 +32,7 @@ export class RealtimeService {
   private listeners: RealtimeListeners = {};
   private sessionTerminatedResolve?: () => void;
 
-  constructor(params: RealtimeServiceParams) {
+  constructor(params: RealtimeTranscriberParams) {
     this.realtimeUrl = params.realtimeUrl ?? defaultRealtimeUrl;
     this.sampleRate = params.sampleRate ?? 16_000;
     this.wordBoost = params.wordBoost;
@@ -157,16 +158,16 @@ export class RealtimeService {
     });
   }
 
-  sendAudio(audio: ArrayBufferLike) {
+  sendAudio(audio: AudioData) {
     if (!this.socket || this.socket.readyState !== WebSocket.OPEN) {
       throw new Error("Socket is not open for communication");
     }
     this.socket.send(audio);
   }
 
-  stream(): WritableStream<ArrayBufferLike> {
-    return new WritableStream<ArrayBufferLike>({
-      write: (chunk: ArrayBufferLike) => {
+  stream(): WritableStream<AudioData> {
+    return new WritableStream<AudioData>({
+      write: (chunk: AudioData) => {
         this.sendAudio(chunk);
       },
     });
@@ -194,3 +195,8 @@ export class RealtimeService {
     this.socket = undefined;
   }
 }
+
+/**
+ * @deprecated Use RealtimeTranscriber instead
+ */
+export class RealtimeService extends RealtimeTranscriber {}

src/types/asyncapi.generated.ts

@@ -15,10 +15,11 @@ type OneOf<T extends any[]> = T extends [infer Only]
   ? OneOf<[XOR<A, B>, ...Rest]>
   : never;
 
-export type AudioData = {
-  /** @description Base64 encoded raw audio data */
-  audio_data: string;
-};
+/**
+ * Format: binary
+ * @description Binary audio data
+ */
+export type AudioData = ArrayBufferLike;
 
 /**
  * @description The encoding of the audio data

src/types/openapi.generated.ts

@@ -475,12 +475,21 @@ export type Error = {
  *     "64nygnr62k-405c-4ae8-8a6b-d90b40ff3cce"
  *   ],
  *   "context": "This is an interview about wildfires.",
+ *   "answer_format": "Bullet Points",
  *   "final_model": "default",
  *   "temperature": 0,
  *   "max_output_size": 3000
  * }
  */
-export type LemurActionItemsParams = LemurBaseParams;
+export type LemurActionItemsParams = LemurBaseParams & {
+  /**
+   * @description How you want the action items to be returned. This can be any text.
+   * Defaults to "Bullet Points".
+   *
+   * @default Bullet Points
+   */
+  answer_format?: string;
+};
 
 /**
  * @example {
@@ -931,6 +940,14 @@ export type RedactedAudioResponse = {
  */
 export type RedactedAudioStatus = "redacted_audio_ready";
 
+/**
+ * @description Controls the filetype of the audio created by redact_pii_audio. Currently supports mp3 (default) and wav. See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more details.
+ * @default mp3
+ * @example mp3
+ * @enum {string}
+ */
+export type RedactPiiAudioQuality = "mp3" | "wav";
+
 /**
  * @example {
  *   "sentences": [
@@ -2076,7 +2093,7 @@ export type Transcript = {
    * @description The audio quality of the PII-redacted audio file, if redact_pii_audio is enabled.
    * See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more information.
    */
-  redact_pii_audio_quality?: string | null;
+  redact_pii_audio_quality?: RedactPiiAudioQuality | null;
   /**
    * @description The list of PII Redaction policies that were enabled, if PII Redaction is enabled.
    * See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more information.
@@ -2244,7 +2261,7 @@ export type TranscriptList = {
  */
 export type TranscriptListItem = {
   audio_url: string;
-  completed?: Date;
+  completed: Date | null;
   created: Date;
   /** Format: uuid */
   id: string;
@@ -2346,7 +2363,7 @@ export type TranscriptOptionalParams = {
    * @description Controls the filetype of the audio created by redact_pii_audio. Currently supports mp3 (default) and wav. See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more details.
    * @default mp3
    */
-  redact_pii_audio_quality?: string;
+  redact_pii_audio_quality?: RedactPiiAudioQuality;
   /** @description The list of PII Redaction policies to enable. See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more details. */
   redact_pii_policies?: PiiPolicy[];
   /** @description The replacement logic for detected PII, can be "entity_type" or "hash". See [PII redaction](https://www.assemblyai.com/docs/models/pii-redaction) for more details. */