feat: add autodownload feature

Author: ChetanXpro

README.md

@@ -1,30 +1,24 @@
-
 # nodejs-whisper
 
 Node.js bindings for OpenAI's Whisper model.
 
-
-
 [![MIT License](https://img.shields.io/badge/License-MIT-green.svg)](https://choosealicense.com/licenses/mit/)
 
-
-
 ## Features
- 
-- Automatically convert the audio to WAV format with a 16000 Hz    frequency to support the whisper model.
-- Output transcripts to (.txt .srt .vtt)
-- Optimized for CPU (Including Apple Silicon ARM)
-- Timestamp precision to single word
-- Split on word rather than on token (Optional)
-- Translate from source language to english (Optional)
-- Convert audio formet to wav to support whisper model
 
+-   Automatically convert the audio to WAV format with a 16000 Hz frequency to support the whisper model.
+-   Output transcripts to (.txt .srt .vtt)
+-   Optimized for CPU (Including Apple Silicon ARM)
+-   Timestamp precision to single word
+-   Split on word rather than on token (Optional)
+-   Translate from source language to english (Optional)
+-   Convert audio formet to wav to support whisper model
 
 ## Installation
 
 1. Install ffmpeg and make tools
 
- ```bash
+```bash
 sudo apt update
 sudo apt install ffmpeg build-essential
 ```
@@ -34,38 +28,39 @@ sudo apt install ffmpeg build-essential
 ```bash
   npm i nodejs-whisper
 ```
+
 2.  Download whisper model
+
 ```bash
   npx nodejs-whisper download
 ```
 
-- NOTE: user may need to install make tool
-    
+-   NOTE: user may need to install make tool
+
 ## Usage/Examples
 
 ```javascript
 import path from 'path'
-import { nodewhisper } from "nodejs-whisper"
+import { nodewhisper } from 'nodejs-whisper'
 
 // Need to provide exact path to your audio file.
 const filePath = path.resolve(__dirname, 'YourAudioFileName')
 
 await nodewhisper(filePath, {
 	modelName: 'base.en', //Downloaded models name
+	autoDownloadModelName: 'base.en', // (optional) autodownload a model if model is not present
 	whisperOptions: {
 		outputInText: false, // get output result in txt file
 		outputInVtt: false, // get output result in vtt file
 		outputInSrt: true, // get output result in srt file
 		outputInCsv: false, // get output result in csv file
-		translateToEnglish: false,//translate from source language to english
+		translateToEnglish: false, //translate from source language to english
 		wordTimestamps: false, // Word-level timestamps
 		timestamps_length: 20, // amount of dialogue per timestamp pair
 		splitOnWord: true, //split on word rather than on token
 	},
 })
 
-
-
 // Model list
 const MODELS_LIST = [
 	'tiny',
@@ -81,15 +76,12 @@ const MODELS_LIST = [
 ]
 ```
 
-
-
-
-
 ## Types
 
 ```
  interface IOptions {
 	modelName: string
+	autoDownloadModelName?: string
 	whisperOptions?: WhisperOptions
 }
 
@@ -106,8 +98,6 @@ const MODELS_LIST = [
 
 ```
 
-
-
 ## Run Locally
 
 Clone the project
@@ -140,17 +130,14 @@ Build Project
   npm run build
 ```
 
-
 ## Made with
 
- - [Whisper OpenAI (using C++ port by: ggerganov)](https://github.com/ggerganov/whisper.cpp)
+-   [Whisper OpenAI (using C++ port by: ggerganov)](https://github.com/ggerganov/whisper.cpp)
 
 ## Feedback
 
 If you have any feedback, please reach out to us at chetanbaliyan10@gmail.com
 
-
 ## Authors
 
-- [@chetanXpro](https://www.github.com/chetanXpro)
-
+-   [@chetanXpro](https://www.github.com/chetanXpro)

package.json

@@ -1,6 +1,6 @@
 {
 	"name": "nodejs-whisper",
-	"version": "0.0.8",
+	"version": "0.1.3",
 	"description": "Node bindings for OpenAI's Whisper. Optimized for CPU.",
 	"main": "dist/index.js",
 	"types": "dist/index.d.ts",

src/autoDownloadModel.ts

@@ -0,0 +1,54 @@
+import path from 'path'
+import shell from 'shelljs'
+
+import { MODELS_LIST, MODELS } from './constants'
+import fs from 'fs'
+
+export default async function autoDownloadModel(autoDownloadModelName?: string) {
+	try {
+		if (autoDownloadModelName) {
+			if (!MODELS_LIST.includes(autoDownloadModelName))
+				throw new Error('[Nodejs-whisper] Error: Provide valid model name')
+
+			shell.cd(path.join(__dirname, '..', './cpp/whisper.cpp/models'))
+
+			let anyModelExist = []
+
+			MODELS.forEach(model => {
+				if (fs.existsSync(path.join(__dirname, '..', `./cpp/whisper.cpp/models/${model}`))) {
+					anyModelExist.push(model)
+					// console.log('anyModelExist found', model)
+				}
+			})
+
+			return new Promise((resolve, reject) => {
+				if (anyModelExist.length > 0) {
+					resolve('Models already exist. Skipping download.')
+
+					// console.log('Models already exist. Skipping download.')
+				} else {
+					console.log(`[Nodejs-whisper] Autodownload Model: ${autoDownloadModelName}\n`)
+
+					let scriptPath = './download-ggml-model.sh'
+
+					if (process.platform === 'win32') scriptPath = 'download-ggml-model.cmd'
+
+					shell.chmod('+x', scriptPath)
+					shell.exec(`${scriptPath} ${autoDownloadModelName}`)
+
+					console.log('[Nodejs-whisper] Attempting to compile model...\n')
+
+					shell.cd('../')
+
+					shell.exec('make')
+
+					resolve('Model Downloaded Successfully')
+				}
+			})
+		}
+	} catch (error) {
+		console.log('[Nodejs-whisper] Error Caught in downloadModel\n')
+		console.log(error)
+		return error
+	}
+}

src/downloadModel.ts

@@ -74,10 +74,9 @@ export default async function downloadModel() {
 		const modelName = await askForModel()
 
 		let scriptPath = './download-ggml-model.sh'
-		// windows .cmd version
+
 		if (process.platform === 'win32') scriptPath = 'download-ggml-model.cmd'
 
-		// todo: check if windows or unix to run bat command or .sh command
 		shell.chmod('+x', scriptPath)
 		shell.exec(`${scriptPath} ${modelName}`)
 

src/index.ts

@@ -1,19 +1,26 @@
 import { WhisperOptions } from './types'
 import { executeCppCommand } from './whisper'
-import downloadModel from './downloadModel'
+// import downloadModel from './downloadModel'
 
 import { constructCommand } from './WhisperHelper'
 import { checkIfFileExists, convertToWavType } from './utils'
 
+import autoDownloadModel from './autoDownloadModel'
+
 export interface IOptions {
 	modelName: string
+	autoDownloadModelName?: string
 	whisperOptions?: WhisperOptions
 }
 
 export async function nodewhisper(filePath: string, options: IOptions) {
+	if (options.autoDownloadModelName) {
+		await autoDownloadModel(options.autoDownloadModelName)
+	}
+
 	checkIfFileExists(filePath)
 	console.log(`[Nodejs-whisper]  Transcribing file: ${filePath}\n`)
-	await downloadModel()
+	// await downloadModel()
 
 	const outputFilePath = await convertToWavType(filePath)
 

src/utils.ts

@@ -20,7 +20,6 @@ export const convertToWavType = async (inputFilePath: string) => {
 	)
 
 	if (fileExtension !== 'wav') {
-		console.warn('[Nodejs-whisper] Warning: Unsupported audio format.\n')
 		console.log('[Nodejs-whisper]  Converting audio to wav File Type...\n')
 		const command = `ffmpeg -nostats -loglevel 0 -i ${inputFilePath} -ar 16000 -ac 1 -c:a pcm_s16le  ${outputFilePath}.wav`
 

src/whisper.ts

@@ -21,7 +21,7 @@ export async function whisperShell(command: string, options: IShellOptions = def
 			// docs: https://github.com/shelljs/shelljs#execcommand--options--callback
 			shell.exec(command, options, (code: number, stdout: string, stderr: string) => {
 				if (code === 0) {
-					console.log('[Nodejs-whisper] Done!')
+					console.log('[Nodejs-whisper] Transcribing Done!')
 
 					resolve(stdout)
 				} else {