Generate SRT from video with OpenAI whisper

getSubtitles.js

const ffmpeg = require("fluent-ffmpeg");
const fs = require("fs");
const path = require("path");
const yargs = require("yargs/yargs");
const { hideBin } = require("yargs/helpers");
const argv = yargs(hideBin(process.argv)).argv;
const { Configuration, OpenAIApi } = require("openai");

// Set OpenAI API key
const configuration = new Configuration({
  apiKey: process.env.OPENAI_API_KEY,
});
const openai = new OpenAIApi(configuration);

// Function to convert MP4 to MP3
function convertToMp3(inputFile, outputFile) {
  return new Promise((resolve, reject) => {
    ffmpeg(inputFile)
      .noVideo()
      .toFormat("mp3")
      .output(outputFile)
      .on("end", resolve)
      .on("error", reject)
      .run();
  });
}

// Function to transcribe audio to SRT
async function transcribeToSrt(inputFile) {
  const resp = await openai.createTranscription(
    fs.createReadStream(inputFile),
    "whisper-1",
    undefined,
    "srt",
    0,
    "en"
  );
  return resp;
}

// Main function to process the video file
async function processVideoFile(videoFilePath) {
  const videoDir = path.dirname(videoFilePath);
  const videoBaseName = path.basename(
    videoFilePath,
    path.extname(videoFilePath)
  );
  const audioFilePath = path.join(videoDir, `${videoBaseName}.mp3`);

  try {
    await convertToMp3(videoFilePath, audioFilePath);
    console.log(`Audio extracted to ${audioFilePath}`);
    const resp = await transcribeToSrt(audioFilePath);
    fs.writeFileSync(`${videoDir}/${videoBaseName}.srt`, resp.data);
    console.log(`SRT file created at ${videoDir}/${videoBaseName}.srt`);
  } catch (err) {
    console.error(`Failed to process video file: ${err}`);
  }
}

// Get the video file path from command line argument
const videoFilePath = argv._[0];
if (!videoFilePath) {
  console.error("Please provide a video file path as a command line argument");
  process.exit(1);
}

processVideoFile(videoFilePath);