Transcribes audio into the input language.

//native

type Groq = {

  api_key: string;

};

type Base64 = string;

/**

 * Transcribes audio into the input language.

 *

 */

export async function main(

  auth: Groq,

  body: {

    file: {

      base64: Base64;

      type:

        | "image/png"

        | "image/jpeg"

        | "image/gif"

        | "application/pdf"

        | "appication/json"

        | "text/csv"

        | "text/plain"

        | "audio/mpeg"

        | "audio/wav"

        | "video/mp4";

      name: string;

    };

    language?: string | false;

    model: string;

    prompt?: string;

    response_format?: "json" | "text" | "verbose_json";

    temperature?: number;

    timestamp_granularities?: "word" | "segment"[];

  },

) {

  const url = new URL(`https://api.groq.com/openai/v1/audio/transcriptions`);

  const formData = new FormData();

  for (const [k, v] of Object.entries(body)) {

    if (v !== undefined && v !== "") {

      if (["file"].includes(k)) {

        const { base64, type, name } = v as {

          base64: Base64;

          type: string;

          name: string;

        };

        formData.append(

          k,

          new Blob([Uint8Array.from(atob(base64), (m) => m.codePointAt(0)!)], {

            type,

          }),

          name,

        );

      } else {

        formData.append(k, String(v));

      }

    }

  }

  const response = await fetch(url, {

    method: "POST",

    headers: {

      Authorization: "Bearer " + auth.api_key,

    },

    body: formData,

  });

  if (!response.ok) {

    const text = await response.text();

    throw new Error(`${response.status} ${text}`);

  }

  return await response.json();

}