@google/genai 2.9.0-rc.0 → 2.9.0

package/dist/vertex_internal/index.d.ts

@@ -322,9 +322,14 @@ declare interface AudioChunk {
 
 /** The audio transcription configuration in Setup. */
 declare interface AudioTranscriptionConfig {
-    /** The language codes of the audio. BCP-47 language code. If not set, the transcription will be in the language detected by the model. If set, the server will use the language code specified in the model config as a hint for the language of the audio
-     */
+    /** Deprecated: use LanguageAuto or LanguageHints instead. */
     languageCodes?: string[];
+    /** The model will detect the language automatically. Do not use together with LanguageHints. */
+    languageAuto?: LanguageAuto;
+    /** Specifies one or more languages in the audio. Do not use together with LanguageAuto. */
+    languageHints?: LanguageHints;
+    /** A list of phrases used for speech adaptation, which biases the ASR model to improve recognition of these specific terms. */
+    adaptationPhrases?: string[];
 }
 
 /**
@@ -4335,6 +4340,16 @@ declare enum Language {
     PYTHON = "PYTHON"
 }
 
+/** Indicates the language of the audio should be automatically detected. */
+declare interface LanguageAuto {
+}
+
+/** Provides hints to the model about possible languages present in the audio. */
+declare interface LanguageHints {
+    /** BCP-47 language codes. At least one must be specified. */
+    languageCodes?: string[];
+}
+
 /** An object that represents a latitude/longitude pair. This is expressed as a pair of doubles to represent degrees latitude and degrees longitude. Unless specified otherwise, this object must conform to the WGS84 standard. Values must be within normalized ranges. */
 declare interface LatLng {
     /** The latitude in degrees. It must be in the range [-90.0, +90.0]. */
@@ -5084,6 +5099,8 @@ declare interface LiveServerContent {
      it is waiting for more input from the user, e.g. because it expects the
      user to continue talking. */
     waitingForInput?: boolean;
+    /** Low latency transcription updated while the user is speaking. */
+    interimInputTranscription?: Transcription;
 }
 
 /** Server will not be able to service client soon. */
@@ -7144,7 +7161,7 @@ declare enum TrafficType {
     PROVISIONED_THROUGHPUT = "PROVISIONED_THROUGHPUT"
 }
 
-/** Audio transcription in Server Conent. */
+/** Audio transcription in Server Content. */
 declare interface Transcription {
     /** Optional. Transcription text. */
     text?: string;
@@ -8081,6 +8098,8 @@ declare namespace types {
         SessionResumptionConfig,
         SlidingWindow,
         ContextWindowCompressionConfig,
+        LanguageAuto,
+        LanguageHints,
         AudioTranscriptionConfig,
         ProactivityConfig,
         CustomizedAvatar,
@@ -8734,6 +8753,8 @@ declare enum VideoOrientation {
 declare interface VoiceActivity {
     /** The type of the voice activity signal. */
     voiceActivityType?: VoiceActivityType;
+    /** The time voice activity detected in audio time, relative to the start of the audio stream. */
+    audioOffset?: string;
 }
 
 declare interface VoiceActivityDetectionSignal {

package/dist/vertex_internal/index.js

@@ -2146,7 +2146,7 @@ const CONTENT_TYPE_HEADER = 'Content-Type';
 const SERVER_TIMEOUT_HEADER = 'X-Server-Timeout';
 const USER_AGENT_HEADER = 'User-Agent';
 const GOOGLE_API_CLIENT_HEADER = 'x-goog-api-client';
-const SDK_VERSION = '2.9.0-rc.0'; // x-release-please-version
+const SDK_VERSION = '2.9.0'; // x-release-please-version
 const LIBRARY_LABEL = `google-genai-sdk/${SDK_VERSION}`;
 const VERTEX_AI_API_DEFAULT_VERSION = 'v1beta1';
 const GOOGLE_AI_API_DEFAULT_VERSION = 'v1beta';