The Resources.Generate class provides methods to generate various types of text content from videos.

Topics, titles, and hashtags

Description: This method generates topics, titles, and hashtags for a specific video. It uses predefined formats and doesn't require a custom prompt, and it's best for generating immediate and straightforward text representations without specific customization.

Function signature and example:

async gist(
  videoId: string,
  types: GenerateGistType[],
  options: RequestOptions = {},
): Promise<Models.GenerateGistResult>

const result = await client.generate.gist(
  "<YOUR_VIDEO_ID>",
  ["title", "topic", "hashtag"]
);

console.log("Result ID:", result.id);

if (result.title !== undefined) {
  console.log("Title:", result.title);
}

if (result.topics !== undefined) {
  console.log("Topics:");
  result.topics.forEach(topic => {
    console.log(`  - ${topic}`);
  });
}

if (result.hashtags !== undefined) {
  console.log("Hashtags:");
  result.hashtags.forEach(hashtag => {
    console.log(`  - ${hashtag}`);
  });
}

Parameters:

Name	Type	Required	Description
`videoId`	`string`	Yes	The unique identifier of the video for which you want to generate text.
`types`	`GenerateGistType[]`	Yes	The types of text you want to generate. Available values: `topics`, `titles`, `hashtags`.
`options`	`RequestOptions`	No	Additional options for the request. Defaults to `{}`.

Return value: Returns a Promise that resolves to a Models.GenerateGistResult instance.

API Reference: For a description of each field in the request and response, see the Generate topics, titles, and hashtags page.

Related guide: Titles, topics, and hashtags.

Summaries, chapters, and highlights

Description: This method generates summaries, chapters, or highlights for your videos. Optionally, you can provide a prompt to customize the output.

Function signature and example:

async summarize(
  videoId: string,
  type: GenerateSummarizeType,
  prompt?: string,
  temperature?: number,
  options: RequestOptions = {},
): Promise<Models.GenerateSummarizeResult>

const result = await client.generate.summarize(
  "<YOUR_VIDEO_ID>",
  "summary",
  "<YOUR_PROMPT>",
  0.7
  
)

console.log(`Result ID: ${result.id}`);

if (result.summary !== undefined) {
  console.log(`Summary: ${result.summary}`);
}

if (result.chapters !== undefined && result.chapters.length > 0) {
  console.log("Chapters:");
  result.chapters.forEach(chapter => {
    console.log(`  Chapter ${chapter.chapterNumber}:`);
    console.log(`    Start: ${chapter.start}`);
    console.log(`    End: ${chapter.end}`);
    console.log(`    Title: ${chapter.chapterTitle}`);
    console.log(`    Summary: ${chapter.chapterSummary}`);
  });
}

if (result.highlights !== undefined && result.highlights.length > 0) {
  console.log("Highlights:");
  result.highlights.forEach(highlight => {
    console.log(`  Start: ${highlight.start}`);
    console.log(`  End: ${highlight.end}`);
    console.log(`  Highlight: ${highlight.highlight}`);
  });
  }

Parameters:

Name	Type	Required	Description
`videoId`	`string`	Yes	The unique identifier of the video for which you want to generate text.
`type`	`string`	Yes	The type of text you want to generate. Available values: `summaries`, `chapters`, `highlights`.
`prompt`	`string`	No	The prompt to customize the output.
`temperature`	`number`	No	The temperature to use in the generation.
`options`	`RequestOptions`	No	Additional options for the request. Defaults to `{}`.

Return value: Returns a Promise that resolves to a Models.GenerateSummarizeResult instance.

API Reference: For a description of each field in the request and response, see the Generate summaries, chapters, and highlights page.

Related guide: Summaries, chapters, and highlights.

Open-ended text

Description: This method generates open-ended texts based on your videos.

Function signature and example:

async text(
  videoId: string,
  prompt: string,
  temperature?: number,
  options: RequestOptions = {},
): Promise<Models.GenerateOpenEndedTextResult>

const result = await client.generate.text(
  "<YOUR_VIDEO_ID>",
  "Be concise"
  0.7,
);
console.log(`Result ID: ${result.id}`);
console.log(`Generated text: ${result.data}`);

Parameters:

Name	Type	Required	Description
`videoId`	`string`	Yes	The unique identifier of the video for which you want to generate text.
`prompt`	`string`	Yes	The prompt to customize the output.
`temperature`	`number`	No	The temperature to use in the generation.
`options`	`RequestOptions`	No	Additional options for the request. Defaults to `{}`.

Return value: Returns a Promise that resolves to a Models.GenerateOpenEndedTextResult instance.

API Reference: For a description of each field in the request and response, see the Open-ended text page.

Related guide: Open-ended text .

Open-ended text with streaming responses

Description: This method generates open-ended texts and supports streaming responses.

Function signature and example:

async textStream(
  { videoId, prompt, temperature }: GenerateTextStreamParams,
  options: RequestOptions = {},
): Promise<Models.GenerateTextStreamResult>

  const textStream = await client.generate.textStream({
    videoId: "<YOUR_VIDEO_ID>",
    prompt: '<YOUR_PROMPT>',
    temperature: 0.7
  });

  for await (const text of textStream) {
    console.log(text);
  }

  console.log(`Aggregated text: ${textStream.aggregatedText}`);
})();

Parameters:

Name	Type	Required	Description
`params`	`GenerateTextStreamParams`	Yes	Parameters for generating open-ended text.
`options`	`RequestOptions`	No	Additional options for the request. Defaults to `{}`.

The GenerateTextStreamParams interface defines the parameters for streaming text generated based on video content:

Name	Type	Required	Description
`videoId`	`string`	Yes	The unique identifier of the video for which you want to generate text.
`prompt`	`string`	Yes	The prompt to customize the output.
`temperature`	`number`	No	The temperature to use in the generation.

API Reference: For a description of each field in the request and response, see the Open-ended text page.

Related guide: Streaming responses .