@defcomm/meetingsdk 0.5.6 → 0.5.8

package/dist/index.d.mts

@@ -143,6 +143,49 @@ type LiveTranscript = {
     end_ms?: number | null;
     timestamp: string;
 };
+type TranscriptSegment = {
+    id: string;
+    transcript_session_id: string;
+    room_id: string;
+    participant_id: string;
+    speaker_name?: string | null;
+    sequence: number;
+    text: string;
+    offset_ms?: number | null;
+    start_ms?: number | null;
+    end_ms?: number | null;
+    created_at: string;
+};
+type TranscriptSession = {
+    id: string;
+    room_id: string;
+    status: string;
+    started_at: string;
+    finalized_at?: string | null;
+};
+type MeetingSummary = {
+    transcript_session_id: string;
+    room_id: string;
+    status: string;
+    overview?: string | null;
+    topics: string[];
+    decisions: string[];
+    action_items: string[];
+    important_moments: Array<{
+        timestamp_ms?: number | null;
+        title?: string | null;
+        description?: string | null;
+    }>;
+    error?: string | null;
+    created_at: string;
+    updated_at: string;
+};
+type TranscriptResponse = {
+    room_id: string;
+    session: TranscriptSession | null;
+    segments: TranscriptSegment[];
+    summary: MeetingSummary | null;
+};
 type NetworkQuality = {
     participant_id: string;
     quality: "excellent" | "good" | "fair" | "poor" | string;
@@ -291,6 +334,10 @@ declare class VideoSDKCore {
     getRoom(roomCode: string): Promise<any>;
     getJoinState(roomCode: string): Promise<JoinState>;
     deleteRoom(roomCode: string): Promise<any>;
+    getTranscript(roomCode: string, sessionId?: string): Promise<TranscriptResponse>;
+    getMeetingSummary(roomCode: string, sessionId: string): Promise<{
+        summary: MeetingSummary | null;
+    }>;
     joinMeeting(config: Omit<MeetingConfig, "participantId">): Promise<void>;
     private connect;
     private createSFUClient;
@@ -368,6 +415,10 @@ type MeetingContextValue = {
     }) => Promise<any>;
     getRoom: (roomCode: string) => Promise<any>;
     deleteRoom: (roomCode: string) => Promise<any>;
+    getTranscript: (roomCode: string, sessionId?: string) => Promise<TranscriptResponse>;
+    getMeetingSummary: (roomCode: string, sessionId: string) => Promise<{
+        summary: MeetingSummary | null;
+    }>;
     toggleMic: () => void;
     toggleCam: () => void;
     startScreenShare: () => Promise<MediaStream | null>;
@@ -453,6 +504,10 @@ declare const useMeeting: (handlers?: {
     }) => Promise<any>;
     getRoom: (roomCode: string) => Promise<any>;
     deleteRoom: (roomCode: string) => Promise<any>;
+    getTranscript: (roomCode: string, sessionId?: string) => Promise<TranscriptResponse>;
+    getMeetingSummary: (roomCode: string, sessionId: string) => Promise<{
+        summary: MeetingSummary | null;
+    }>;
     toggleMic: () => void;
     toggleCam: () => void;
     startScreenShare: () => Promise<MediaStream | null>;
@@ -590,4 +645,4 @@ declare function usePubSub<T extends PubSubTopic>(topic: T, options?: {
     onMessageReceived?: (message: PubSubMessage<PubSubTopicConfig[T]["message"]>) => void;
 }): PubSubHandleForTopic<T>;
 
-export { ActiveSpeakerDetector, type ChatInput, type LiveTranscript, type MeetingConfig, MeetingProvider, MeetingState, type NetworkQuality, type Participant, type ParticipantInfo, type PubSubHandle, type PubSubHandleForTopic, type PubSubMessage, type PubSubTopic, type PubSubTopicConfig, type Reaction, type RecordingInfo, type SFUConnectionState, type TrackInfo, VideoSDKCore, useLiveTranscript, useLocalParticipant, useMeeting, useMeetingContext, useMeetingPreview, useNetworkQuality, useParticipants, usePubSub, useRemoteMedia };
+export { ActiveSpeakerDetector, type ChatInput, type LiveTranscript, type MeetingConfig, MeetingProvider, MeetingState, type MeetingSummary, type NetworkQuality, type Participant, type ParticipantInfo, type PubSubHandle, type PubSubHandleForTopic, type PubSubMessage, type PubSubTopic, type PubSubTopicConfig, type Reaction, type RecordingInfo, type SFUConnectionState, type TrackInfo, type TranscriptResponse, type TranscriptSegment, type TranscriptSession, VideoSDKCore, useLiveTranscript, useLocalParticipant, useMeeting, useMeetingContext, useMeetingPreview, useNetworkQuality, useParticipants, usePubSub, useRemoteMedia };

package/dist/index.d.ts

@@ -143,6 +143,49 @@ type LiveTranscript = {
     end_ms?: number | null;
     timestamp: string;
 };
+type TranscriptSegment = {
+    id: string;
+    transcript_session_id: string;
+    room_id: string;
+    participant_id: string;
+    speaker_name?: string | null;
+    sequence: number;
+    text: string;
+    offset_ms?: number | null;
+    start_ms?: number | null;
+    end_ms?: number | null;
+    created_at: string;
+};
+type TranscriptSession = {
+    id: string;
+    room_id: string;
+    status: string;
+    started_at: string;
+    finalized_at?: string | null;
+};
+type MeetingSummary = {
+    transcript_session_id: string;
+    room_id: string;
+    status: string;
+    overview?: string | null;
+    topics: string[];
+    decisions: string[];
+    action_items: string[];
+    important_moments: Array<{
+        timestamp_ms?: number | null;
+        title?: string | null;
+        description?: string | null;
+    }>;
+    error?: string | null;
+    created_at: string;
+    updated_at: string;
+};
+type TranscriptResponse = {
+    room_id: string;
+    session: TranscriptSession | null;
+    segments: TranscriptSegment[];
+    summary: MeetingSummary | null;
+};
 type NetworkQuality = {
     participant_id: string;
     quality: "excellent" | "good" | "fair" | "poor" | string;
@@ -291,6 +334,10 @@ declare class VideoSDKCore {
     getRoom(roomCode: string): Promise<any>;
     getJoinState(roomCode: string): Promise<JoinState>;
     deleteRoom(roomCode: string): Promise<any>;
+    getTranscript(roomCode: string, sessionId?: string): Promise<TranscriptResponse>;
+    getMeetingSummary(roomCode: string, sessionId: string): Promise<{
+        summary: MeetingSummary | null;
+    }>;
     joinMeeting(config: Omit<MeetingConfig, "participantId">): Promise<void>;
     private connect;
     private createSFUClient;
@@ -368,6 +415,10 @@ type MeetingContextValue = {
     }) => Promise<any>;
     getRoom: (roomCode: string) => Promise<any>;
     deleteRoom: (roomCode: string) => Promise<any>;
+    getTranscript: (roomCode: string, sessionId?: string) => Promise<TranscriptResponse>;
+    getMeetingSummary: (roomCode: string, sessionId: string) => Promise<{
+        summary: MeetingSummary | null;
+    }>;
     toggleMic: () => void;
     toggleCam: () => void;
     startScreenShare: () => Promise<MediaStream | null>;
@@ -453,6 +504,10 @@ declare const useMeeting: (handlers?: {
     }) => Promise<any>;
     getRoom: (roomCode: string) => Promise<any>;
     deleteRoom: (roomCode: string) => Promise<any>;
+    getTranscript: (roomCode: string, sessionId?: string) => Promise<TranscriptResponse>;
+    getMeetingSummary: (roomCode: string, sessionId: string) => Promise<{
+        summary: MeetingSummary | null;
+    }>;
     toggleMic: () => void;
     toggleCam: () => void;
     startScreenShare: () => Promise<MediaStream | null>;
@@ -590,4 +645,4 @@ declare function usePubSub<T extends PubSubTopic>(topic: T, options?: {
     onMessageReceived?: (message: PubSubMessage<PubSubTopicConfig[T]["message"]>) => void;
 }): PubSubHandleForTopic<T>;
 
-export { ActiveSpeakerDetector, type ChatInput, type LiveTranscript, type MeetingConfig, MeetingProvider, MeetingState, type NetworkQuality, type Participant, type ParticipantInfo, type PubSubHandle, type PubSubHandleForTopic, type PubSubMessage, type PubSubTopic, type PubSubTopicConfig, type Reaction, type RecordingInfo, type SFUConnectionState, type TrackInfo, VideoSDKCore, useLiveTranscript, useLocalParticipant, useMeeting, useMeetingContext, useMeetingPreview, useNetworkQuality, useParticipants, usePubSub, useRemoteMedia };
+export { ActiveSpeakerDetector, type ChatInput, type LiveTranscript, type MeetingConfig, MeetingProvider, MeetingState, type MeetingSummary, type NetworkQuality, type Participant, type ParticipantInfo, type PubSubHandle, type PubSubHandleForTopic, type PubSubMessage, type PubSubTopic, type PubSubTopicConfig, type Reaction, type RecordingInfo, type SFUConnectionState, type TrackInfo, type TranscriptResponse, type TranscriptSegment, type TranscriptSession, VideoSDKCore, useLiveTranscript, useLocalParticipant, useMeeting, useMeetingContext, useMeetingPreview, useNetworkQuality, useParticipants, usePubSub, useRemoteMedia };

package/dist/index.js

@@ -472,11 +472,7 @@ var SFUClient = class {
     this.trackToSource = /* @__PURE__ */ new Map();
     this.localTrackIds = /* @__PURE__ */ new Map();
     this.subscriberStatsTimer = null;
-    this.liveTranscriptionActive = false;
-    this.liveTranscriptionSequence = 0;
-    this.liveTranscriptionStartedAt = 0;
-    this.liveTranscriptionRecorder = null;
-    this.liveTranscriptionLoopPromise = null;
+    this.liveTranscriptionSeen = false;
     this.networkProfile = null;
     this.heartbeatTimer = null;
     this.apiBase = options.apiBase || "https://api.example.com/api";
@@ -1144,7 +1140,10 @@ var SFUClient = class {
           end_ms: payload.end_ms == null ? null : Number(payload.end_ms),
           timestamp: String(payload.timestamp ?? (/* @__PURE__ */ new Date()).toISOString())
         };
-        if (transcript.text.trim()) this.onLiveTranscript(transcript);
+        if (transcript.text.trim()) {
+          this.liveTranscriptionSeen = true;
+          this.onLiveTranscript(transcript);
+        }
         break;
       }
       case "network_quality": {
@@ -1825,119 +1824,63 @@ var SFUClient = class {
       }
     });
   }
-  getLiveTranscriptionMimeType() {
-    if (typeof MediaRecorder === "undefined") {
-      throw new Error("MediaRecorder is not supported by this browser");
+  /**
+   * Transcription is server-authoritative. The SFU reads the already-published
+   * microphone RTP stream, sends audio to the configured transcription service,
+   * persists final segments, and broadcasts LiveTranscript events.
+   *
+   * These methods remain as compatibility shims for applications that used the
+   * previous browser MediaRecorder-based API. They intentionally do not capture
+   * or upload microphone audio from the browser.
+   */
+  async startLiveTranscription() {
+    if (!this.room) {
+      throw new Error("Cannot start live transcription without a room");
     }
-    const candidates = [
-      "audio/webm;codecs=opus",
-      "audio/webm",
-      "audio/ogg;codecs=opus"
-    ];
-    return candidates.find((type) => MediaRecorder.isTypeSupported(type)) || "";
-  }
-  async startLiveTranscription(audioTrack, speakerName) {
-    if (this.liveTranscriptionActive) return;
-    if (!this.room) throw new Error("Cannot start live transcription without a room");
-    if (!this.authToken) throw new Error("Cannot start live transcription without an authenticated token");
-    if (!audioTrack || audioTrack.kind !== "audio") {
-      throw new Error("A live microphone audio track is required");
-    }
-    const mimeType = this.getLiveTranscriptionMimeType();
-    this.liveTranscriptionActive = true;
-    this.liveTranscriptionSequence = 0;
-    this.liveTranscriptionStartedAt = Date.now();
-    const stream = new MediaStream([audioTrack]);
-    this.liveTranscriptionLoopPromise = this.runLiveTranscriptionLoop(
-      stream,
-      mimeType
-    ).finally(() => {
-      this.liveTranscriptionRecorder = null;
-      this.liveTranscriptionLoopPromise = null;
-    });
-    void speakerName;
-    await Promise.resolve();
+    console.info("[Transcript] server-authoritative live transcription is managed by the SFU");
   }
-  async captureLiveChunk(stream, mimeType) {
-    return new Promise((resolve, reject) => {
-      const chunks = [];
-      const recorder = mimeType ? new MediaRecorder(stream, { mimeType }) : new MediaRecorder(stream);
-      this.liveTranscriptionRecorder = recorder;
-      const timeout = setTimeout(() => {
-        if (recorder.state !== "inactive") recorder.stop();
-      }, 4500);
-      recorder.ondataavailable = (event) => {
-        if (event.data.size > 0) chunks.push(event.data);
-      };
-      recorder.onerror = () => {
-        clearTimeout(timeout);
-        reject(new Error("Live transcription recorder failed"));
-      };
-      recorder.onstop = () => {
-        clearTimeout(timeout);
-        this.liveTranscriptionRecorder = null;
-        resolve(
-          chunks.length ? new Blob(chunks, { type: recorder.mimeType || mimeType || "audio/webm" }) : null
-        );
-      };
-      recorder.start();
-    });
+  stopLiveTranscription() {
+    console.info("[Transcript] server-authoritative live transcription remains managed by the SFU");
   }
-  async runLiveTranscriptionLoop(stream, mimeType) {
-    while (this.liveTranscriptionActive && stream.getAudioTracks()[0]?.readyState === "live") {
-      try {
-        const blob = await this.captureLiveChunk(stream, mimeType);
-        if (!this.liveTranscriptionActive) break;
-        if (!blob || blob.size === 0) continue;
-        const sequence = this.liveTranscriptionSequence++;
-        const offsetMs = Math.max(0, Date.now() - this.liveTranscriptionStartedAt);
-        await this.uploadLiveTranscriptionChunk(blob, sequence, offsetMs, mimeType);
-      } catch (error) {
-        this.onError(error);
-        await new Promise((resolve) => setTimeout(resolve, 750));
-      }
-    }
+  isLiveTranscriptionActive() {
+    return this.liveTranscriptionSeen;
   }
-  async uploadLiveTranscriptionChunk(blob, sequence, offsetMs, mimeType) {
-    if (!this.room || !this.authToken) return;
-    const form = new FormData();
-    form.append("sequence", String(sequence));
-    form.append("offset_ms", String(offsetMs));
-    const extension = mimeType.includes("ogg") || blob.type.includes("ogg") ? "ogg" : "webm";
-    form.append("audio", blob, `live-${sequence}.${extension}`);
+  async getTranscript(roomCode, sessionId) {
+    if (!this.authToken) throw new Error("Authentication token is required");
+    const query = sessionId ? `?session_id=${encodeURIComponent(sessionId)}` : "";
     const response = await fetch(
-      `${this.apiBase}/rooms/${encodeURIComponent(this.room)}/live-transcription`,
+      `${this.apiBase}/rooms/${encodeURIComponent(roomCode)}/transcript${query}`,
       {
-        method: "POST",
+        method: "GET",
         headers: {
           Authorization: `Bearer ${this.authToken}`,
           Accept: "application/json"
-        },
-        body: form
+        }
       }
     );
+    const text = await response.text();
     if (!response.ok) {
-      let detail = `Live transcription request failed: ${response.status}`;
-      try {
-        const body = await response.json();
-        detail = body?.message || body?.detail || detail;
-      } catch {
-      }
-      throw new Error(detail);
+      throw new Error(`Transcript request failed: ${response.status} ${text}`);
     }
+    return JSON.parse(text);
   }
-  stopLiveTranscription() {
-    this.liveTranscriptionActive = false;
-    try {
-      if (this.liveTranscriptionRecorder?.state !== "inactive") {
-        this.liveTranscriptionRecorder?.stop();
+  async getMeetingSummary(roomCode, sessionId) {
+    if (!this.authToken) throw new Error("Authentication token is required");
+    const response = await fetch(
+      `${this.apiBase}/rooms/${encodeURIComponent(roomCode)}/transcript/${encodeURIComponent(sessionId)}/summary`,
+      {
+        method: "GET",
+        headers: {
+          Authorization: `Bearer ${this.authToken}`,
+          Accept: "application/json"
+        }
       }
-    } catch {
+    );
+    const text = await response.text();
+    if (!response.ok) {
+      throw new Error(`Meeting summary request failed: ${response.status} ${text}`);
     }
-    this.liveTranscriptionRecorder = null;
-  }
-  isLiveTranscriptionActive() {
-    return this.liveTranscriptionActive;
+    return JSON.parse(text);
   }
   async applyNetworkRecommendation(action) {
     if (!this.videoSender || this.networkProfile === action) return;
@@ -2380,6 +2323,12 @@ var VideoSDKCore = class {
   async deleteRoom(roomCode) {
     return this.sfuClient.deleteRoom(roomCode);
   }
+  async getTranscript(roomCode, sessionId) {
+    return this.sfuClient.getTranscript(roomCode, sessionId);
+  }
+  async getMeetingSummary(roomCode, sessionId) {
+    return this.sfuClient.getMeetingSummary(roomCode, sessionId);
+  }
   // JOIN
   async joinMeeting(config) {
     const { roomId, name, audioMuted = false, videoMuted = false } = config;
@@ -3164,13 +3113,10 @@ var VideoSDKCore = class {
       throw new Error("SFU client is not connected");
     }
     const audioTrack = this.localStream?.getAudioTracks()[0];
-    if (!audioTrack) {
+    if (!audioTrack || audioTrack.readyState !== "live") {
       throw new Error("Microphone audio track is not available");
     }
-    await this.sfuClient.startLiveTranscription(
-      audioTrack,
-      this.participantName
-    );
+    await this.sfuClient.startLiveTranscription();
   }
   stopLiveTranscription() {
     this.sfuClient?.stopLiveTranscription();
@@ -3436,6 +3382,8 @@ var MeetingProvider = ({
       createRoom: sdk.createRoom.bind(sdk),
       getRoom: sdk.getRoom.bind(sdk),
       deleteRoom: sdk.deleteRoom.bind(sdk),
+      getTranscript: sdk.getTranscript.bind(sdk),
+      getMeetingSummary: sdk.getMeetingSummary.bind(sdk),
       toggleMic: sdk.toggleMic.bind(sdk),
       toggleCam: sdk.toggleCam.bind(sdk),
       startScreenShare: sdk.startScreenShare.bind(sdk),