implemented method for generating captions from a clip

Author: k9p5

playground/main.ts

@@ -75,13 +75,13 @@ await composition.add(
     start: composition.findClips(core.VideoClip).at(0)?.stop.subtract(new core.Timestamp(6000)),
     stop: composition.duration
   })
-)
+);
 
-await composition.add(
+(await composition.add(
   new core.AudioClip(await core.AudioSource.from('/audio.mp3'), {
     transcript: core.Transcript.fromJSON(captions).optimize(),
   })
-);
+)).generateCaptions();
 
 await composition.add(
   new core.TextClip({
@@ -130,7 +130,3 @@ await composition.add(
     }]
   })
 );
-
-await composition.createTrack('caption')
-  .from(composition.findClips(core.AudioClip)[0])
-  .create();

src/clips/media/media.spec.ts

@@ -9,12 +9,14 @@ import { describe, expect, it, vi, beforeEach, afterAll } from 'vitest';
 import { Keyframe, Timestamp, Transcript } from '../../models';
 import { captions } from '../../test/captions';
 import { Composition } from '../../composition';
-import { MediaTrack } from '../../tracks';
+import { CaptionTrack, MediaTrack } from '../../tracks';
+import { MediaClipProps } from './media.interfaces';
+import { Font, TextClip } from '../text';
+import { ValidationError } from '../../errors';
 import { VisualMixin, VisualMixinProps } from '../mixins';
 import { MediaClip } from './media';
 
 import type { frame, MixinType } from '../../types';
-import { MediaClipProps } from './media.interfaces';
 
 describe('The Media Clip', () => {
 	const mockFn = vi.fn();
@@ -285,6 +287,34 @@ describe('The Media Clip', () => {
 		expect(copy.range[1]).toBeInstanceOf(Timestamp);
 		expect(copy.range[1].frames).toBe(80);
 	});
+
+	it('should generate a caption track using a transcript', async () => {
+		vi.spyOn(Font.prototype, 'load').mockImplementation(async () => new Font());
+		clip.transcript = Transcript.fromJSON(captions);
+		clip.state = 'READY';
+
+		const composition = new Composition();
+		await composition.add(clip);
+
+		await clip.generateCaptions();
+
+		expect(composition.tracks.length).toBe(2);
+
+		const track = composition.tracks[0];
+
+		expect(track).toBeInstanceOf(CaptionTrack);
+		expect(track.clips.length).toBe(36);
+		expect(track.clips[0]).toBeInstanceOf(TextClip);
+		expect((track.clips[0] as TextClip).text).toBe('Is the');
+	});
+
+	it('should throw an error when trying to generate captions without a composition', async () => {
+		vi.spyOn(Font.prototype, 'load').mockImplementation(async () => new Font());
+		clip.transcript = Transcript.fromJSON(captions);
+		clip.state = 'READY';
+
+		await expect(() => clip.generateCaptions()).rejects.toThrowError(ValidationError)
+	});
 });
 
 describe('Split tests - the Media Clip object', () => {

src/clips/media/media.ts

@@ -11,9 +11,10 @@ import { AsyncMixin } from '../mixins';
 import { RangeDeserializer } from './media.deserializer';
 import { serializable, } from '../../services';
 import { replaceKeyframes } from '../clip/clip.utils';
+import { ValidationError } from '../../errors';
 import { Clip } from '../clip';
 
-import type { Track } from '../../tracks';
+import type { CaptionPresetStrategy, Track } from '../../tracks';
 import type { float, frame } from '../../types';
 import type { MediaClipProps } from './media.interfaces';
 
@@ -274,6 +275,27 @@ export class MediaClip<Props extends MediaClipProps = MediaClipProps> extends As
 		return copy;
 	}
 
+	/**
+	 * Generates a new caption track for the current clip using the specified captioning strategy.
+	 * @param preset An optional CaptionPresetStrategy to define how captions should be generated.
+	 * @returns {Promise<void>} A promise that resolves when the caption track has been successfully created.
+	 */
+	public async generateCaptions(preset?: CaptionPresetStrategy): Promise<this> {
+		if (!this.track?.composition) {
+			throw new ValidationError({
+				i18n: 'compositionNotDefined',
+				message: 'Captions can only be generated after the clip has been added to the composition',
+			});
+		}
+
+		await this.track.composition
+			.createTrack('caption')
+			.from(this)
+			.create(preset);
+
+		return this;
+	}
+
 	public set(props?: Props): this {
 		return super.set(props);
 	}