Skip to content

Publish transcriptions additionally via text stream APIs #348

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 4 commits into
base: main
Choose a base branch
from
Open

Publish transcriptions additionally via text stream APIs #348

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
703f808
publish transcriptions additionally via text stream APIs
lukasIO Mar 28, 2025
693c298
align publishTranscription signature
lukasIO Mar 28, 2025
6c36cf7
cleanup
lukasIO Mar 28, 2025
dacf328
reuse
lukasIO Mar 28, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 7 additions & 0 deletions agents/src/constants.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,7 @@
// SPDX-FileCopyrightText: 2024 LiveKit, Inc.
//
// SPDX-License-Identifier: Apache-2.0
export const ATTRIBUTE_TRANSCRIPTION_TRACK_ID = 'lk.transcribed_track_id';
export const ATTRIBUTE_TRANSCRIPTION_FINAL = 'lk.transcription_final';
export const TOPIC_TRANSCRIPTION = 'lk.transcription';
export const TOPIC_CHAT = 'lk.chat';
42 changes: 32 additions & 10 deletions agents/src/multimodal/multimodal_agent.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,6 +19,11 @@ import {
} from '@livekit/rtc-node';
import { EventEmitter } from 'node:events';
import { AudioByteStream } from '../audio.js';
import {
ATTRIBUTE_TRANSCRIPTION_FINAL,
ATTRIBUTE_TRANSCRIPTION_TRACK_ID,
TOPIC_TRANSCRIPTION,
} from '../constants.js';
import * as llm from '../llm/index.js';
import { log } from '../log.js';
import type { MultimodalLLMMetrics } from '../metrics/base.js';
Expand Down Expand Up @@ -246,8 +251,8 @@ export class MultimodalAgent extends EventEmitter {
if (message.contentType === 'text') return;

const synchronizer = new TextAudioSynchronizer(defaultTextSyncOptions);
synchronizer.on('textUpdated', (text) => {
this.#publishTranscription(
synchronizer.on('textUpdated', async (text) => {
await this.#publishTranscription(
this.room!.localParticipant!.identity!,
this.#getLocalTrackSid()!,
text.text,
Expand Down Expand Up @@ -297,25 +302,31 @@ export class MultimodalAgent extends EventEmitter {
});

// eslint-disable-next-line @typescript-eslint/no-explicit-any
this.#session.on('input_speech_committed', (ev: any) => {
this.#session.on('input_speech_committed', async (ev: any) => {
// openai.realtime.InputSpeechCommittedEvent
const participantIdentity = this.linkedParticipant?.identity;
const trackSid = this.subscribedTrack?.sid;
if (participantIdentity && trackSid) {
this.#publishTranscription(participantIdentity, trackSid, '…', false, ev.itemId);
await this.#publishTranscription(participantIdentity, trackSid, '…', false, ev.itemId);
} else {
this.#logger.error('Participant or track not set');
}
});

// eslint-disable-next-line @typescript-eslint/no-explicit-any
this.#session.on('input_speech_transcription_completed', (ev: any) => {
this.#session.on('input_speech_transcription_completed', async (ev: any) => {
// openai.realtime.InputSpeechTranscriptionCompletedEvent
const transcription = ev.transcript;
const participantIdentity = this.linkedParticipant?.identity;
const trackSid = this.subscribedTrack?.sid;
if (participantIdentity && trackSid) {
this.#publishTranscription(participantIdentity, trackSid, transcription, true, ev.itemId);
await this.#publishTranscription(
participantIdentity,
trackSid,
transcription,
true,
ev.itemId,
);
} else {
this.#logger.error('Participant or track not set');
}
Expand All @@ -327,7 +338,7 @@ export class MultimodalAgent extends EventEmitter {
this.#logger.child({ transcription }).debug('committed user speech');
});

this.#session.on('input_speech_started', (ev: any) => {
this.#session.on('input_speech_started', async (ev: any) => {
this.emit('user_started_speaking');
if (this.#playingHandle && !this.#playingHandle.done) {
this.#playingHandle.interrupt();
Expand All @@ -344,7 +355,7 @@ export class MultimodalAgent extends EventEmitter {
const participantIdentity = this.linkedParticipant?.identity;
const trackSid = this.subscribedTrack?.sid;
if (participantIdentity && trackSid) {
this.#publishTranscription(participantIdentity, trackSid, '…', false, ev.itemId);
await this.#publishTranscription(participantIdentity, trackSid, '…', false, ev.itemId);
}
});

Expand Down Expand Up @@ -467,13 +478,13 @@ export class MultimodalAgent extends EventEmitter {
return this.#localTrackSid;
}

#publishTranscription(
async #publishTranscription(
participantIdentity: string,
trackSid: string,
text: string,
isFinal: boolean,
id: string,
): void {
): Promise<void> {
this.#logger.debug(
`Publishing transcription ${participantIdentity} ${trackSid} ${text} ${isFinal} ${id}`,
);
Expand All @@ -496,6 +507,17 @@ export class MultimodalAgent extends EventEmitter {
},
],
});

const stream = await this.room.localParticipant.streamText({
topic: TOPIC_TRANSCRIPTION,
senderIdentity: participantIdentity,
attributes: {
[ATTRIBUTE_TRANSCRIPTION_TRACK_ID]: trackSid,
[ATTRIBUTE_TRANSCRIPTION_FINAL]: isFinal.toString(),
},
});
await stream.write(text);
await stream.close();
}

#updateState() {
Expand Down
100 changes: 64 additions & 36 deletions agents/src/pipeline/pipeline_agent.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,6 +12,11 @@ import {
import type { TypedEventEmitter as TypedEmitter } from '@livekit/typed-emitter';
import { randomUUID } from 'node:crypto';
import EventEmitter from 'node:events';
import {
ATTRIBUTE_TRANSCRIPTION_FINAL,
ATTRIBUTE_TRANSCRIPTION_TRACK_ID,
TOPIC_TRANSCRIPTION,
} from '../constants.js';
import type {
CallableFunctionResult,
FunctionCallInfo,
Expand Down Expand Up @@ -505,28 +510,21 @@ export class VoicePipelineAgent extends (EventEmitter as new () => TypedEmitter<
this.emit(VPAEvent.USER_STOPPED_SPEAKING);
this.#deferredValidation.onHumanEndOfSpeech(event);
});
this.#humanInput.on(HumanInputEvent.INTERIM_TRANSCRIPT, (event) => {
this.#humanInput.on(HumanInputEvent.INTERIM_TRANSCRIPT, async (event) => {
if (!this.#transcriptionId) {
this.#transcriptionId = randomUUID();
}
this.#transcribedInterimText = event.alternatives![0].text;

this.#room!.localParticipant!.publishTranscription({
participantIdentity: this.#humanInput!.participant.identity,
trackSid: this.#humanInput!.subscribedTrack!.sid!,
segments: [
{
text: this.#transcribedInterimText,
id: this.#transcriptionId,
final: true,
startTime: BigInt(0),
endTime: BigInt(0),
language: '',
},
],
});
await this.#publishTranscription(
this.#humanInput!.participant.identity,
this.#humanInput!.subscribedTrack!.sid!,
this.#transcribedInterimText,
false,
this.#transcriptionId,
);
});
this.#humanInput.on(HumanInputEvent.FINAL_TRANSCRIPT, (event) => {
this.#humanInput.on(HumanInputEvent.FINAL_TRANSCRIPT, async (event) => {
const newTranscript = event.alternatives![0].text;
if (!newTranscript) return;

Expand All @@ -537,20 +535,14 @@ export class VoicePipelineAgent extends (EventEmitter as new () => TypedEmitter<
this.#lastFinalTranscriptTime = Date.now();
this.transcribedText += (this.transcribedText ? ' ' : '') + newTranscript;

this.#room!.localParticipant!.publishTranscription({
participantIdentity: this.#humanInput!.participant.identity,
trackSid: this.#humanInput!.subscribedTrack!.sid!,
segments: [
{
text: this.transcribedText,
id: this.#transcriptionId,
final: true,
startTime: BigInt(0),
endTime: BigInt(0),
language: '',
},
],
});
await this.#publishTranscription(
this.#humanInput!.participant.identity,
this.#humanInput!.subscribedTrack!.sid!,
this.transcribedText,
true,
this.#transcriptionId,
);

this.#transcriptionId = undefined;

if (
Expand Down Expand Up @@ -881,18 +873,54 @@ export class VoicePipelineAgent extends (EventEmitter as new () => TypedEmitter<
handle.setDone();
}

async #publishTranscription(
participantIdentity: string,
trackSid: string,
text: string,
isFinal: boolean,
id: string,
) {
this.#room!.localParticipant!.publishTranscription({
participantIdentity: participantIdentity,
trackSid: trackSid,
segments: [
{
text: text,
final: isFinal,
id: id,
startTime: BigInt(0),
endTime: BigInt(0),
language: '',
},
],
});
const stream = await this.#room!.localParticipant!.streamText({
senderIdentity: participantIdentity,
topic: TOPIC_TRANSCRIPTION,
attributes: {
[ATTRIBUTE_TRANSCRIPTION_TRACK_ID]: trackSid,
[ATTRIBUTE_TRANSCRIPTION_FINAL]: isFinal.toString(),
},
});
await stream.write(text);
await stream.close();
}

#synthesizeAgentSpeech(
speechId: string,
source: string | LLMStream | AsyncIterable<string>,
): SynthesisHandle {
const synchronizer = new TextAudioSynchronizer(defaultTextSyncOptions);
synchronizer.on('textUpdated', (text) => {
// TODO: where possible we would want to use deltas instead of full text segments, esp for LLM streams over the streamText API
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this makes sense.

synchronizer.on('textUpdated', async (text) => {
this.#agentTranscribedText = text.text;
this.#room!.localParticipant!.publishTranscription({
participantIdentity: this.#room!.localParticipant!.identity,
trackSid: this.#agentPublication!.sid!,
segments: [text],
});
await this.#publishTranscription(
this.#room!.localParticipant!.identity!,
this.#agentPublication?.sid ?? '',
text.text,
text.final,
text.id,
);
});

if (!this.#agentOutput) {
Expand Down
Loading