# Auto detect text files and perform LF normalization
* text=auto


.gitattributes

node_modules
storage/datasets/default/*
.env
# Added by Apify CLI
storage
.venv
.output

.gitignore

single-tweet-scraper.apify

package.json

tsconfig.json

# Specify the base Docker image. You can read more about
# the available images at https://sdk.apify.com/docs/guides/docker-images
# You can also use any other image from Docker Hub.
FROM apify/actor-node:16

# Copy just package.json and package-lock.json
# to speed up the build using Docker layer cache.
COPY package*.json ./

# Install NPM packages, skip optional and development dependencies to
# keep the image small. Avoid logging too much and print the dependency
# tree for debugging
RUN npm --quiet set progress=false \
    && npm install --omit=dev --omit=optional \
    && echo "Installed NPM packages:" \
    && (npm list --omit=dev --all || true) \
    && echo "Node.js version:" \
    && node --version \
    && echo "NPM version:" \
    && npm --version \
    && rm -r ~/.npm

# Next, copy the remaining files and directories with the source code.
# Since we do this after NPM install, quick build will be really fast
# for most source file changes.
COPY . ./


# Run the image.
CMD npm start --silent


.actor/Dockerfile

# Speech Text Actor

This actor helps you to get transcription for any audio files in seconds.

## Use Cases

You need to provide URL of audio file and choose the input format and output format you want to get. The actor will return you the transcription in the output format you choose.

## Input

The URL format should be string. You can upload it somewhere temporarily and provide the downlad URL. The downloaded file will be removed after the transcription is done and won't be stored anywhere.

The input format can be one of the following:

| Format | Description |
| ------ | ----------- |
| flac | FLAC audio file |
| m4a | M4A audio file |
| mp3 | MP3 audio file |
| mp4 | MP4 audio file |
| mpeg | MPEG audio file |
| mpga | MPGA audio file |
| oga | OGA audio file |
| ogg | OGG audio file |
| wav | WAV audio file |
| webm | WEBM audio file |


## Output

The output format can be one of the following:

| Format | Description |
| ------ | ----------- |
| json | JSON format |
| text | Plain text format |
| srt | SRT subtitle format |
| verbose_json | Verbose JSON format |
| vtt | VTT subtitle format |

.actor/README.md

speech-apify

Speech To Text

.actor/actor.json

File

Input Format

Output Format

Actor input schema

.actor/input-schema.json

Transcription Output

Overview

.actor/output-schema.json

import { Actor } from 'apify';
import fetch from 'node-fetch';
import fs from 'fs';
import FormData from 'form-data';
import dotenv from 'dotenv';
dotenv.config();
// Input configuration and controls
await Actor.init();
const input = await Actor.getInput();
if (!input?.file)
    throw new Error('File URL is missing.');
const downloadFile = async (url) => {
    return fetch(url)
        .then(res => {
        const dest = fs.createWriteStream('./output');
        res.body.pipe(dest);
        return new Promise((resolve, reject) => {
            dest.on('finish', resolve);
            dest.on('error', reject);
        });
    });
};
const produceCaption = async (file, options) => {
    var formdata = new FormData();
    formdata.append("file", file.buffer, file.name);
    formdata.append("model", options.model);
    formdata.append("response_format", options.response_format);
    formdata.append("prompt", "Each part shouldn't be longer than 3 seconds. Each part should have only one sentence and shouldn't have uncompleted sentence.");
    var requestOptions = {
        method: 'POST',
        headers: { "Authorization": `Bearer ${process.env.OPENAI_API_KEY}` },
        body: formdata,
        redirect: 'follow'
    };
    const response = await fetch("https://api.openai.com/v1/audio/transcriptions", requestOptions);
    return {
        success: response.status === 200,
        body: response.status === 200 ? await response.text() : await response.json()
    };
};
const inputFileExtension = {
    "flac": ".flac",
    "m4a": ".m4a",
    "mp3": ".mp3",
    "mp4": ".mp4",
    "mpeg": ".mpeg",
    "mpga": ".mpga",
    "oga": ".oga",
    "ogg": ".ogg",
    "wav": ".wav",
    "webm": ".webm"
};
await downloadFile(input.file);
const fileReadStream = fs.createReadStream(`./output`);
const captionResponse = await produceCaption({
    buffer: fileReadStream, name: 'output' + inputFileExtension[input.inputFormat]
}, {
    model: 'whisper-1', response_format: input.outputFormat
});
fs.unlinkSync(`./output`);
await Actor.pushData({
    audioUrl: input.file,
    inputFormat: input.inputFormat,
    outputFormat: input.outputFormat,
    outputText: captionResponse.body
});
await Actor.exit();
await Actor.exit();


build/main.js

import { Actor } from 'apify'
import { IInput } from './types';
import fetch from 'node-fetch';
import fs from 'fs';
import FormData from 'form-data';
import { Stream } from 'stream';

import dotenv from 'dotenv';
dotenv.config();

// Input configuration and controls
await Actor.init();
const input: IInput = await Actor.getInput();
if (!input?.file) throw new Error('File URL is missing.');

const downloadFile = async (url: string) => {
    return fetch(url)
        .then(res => {
            const dest = fs.createWriteStream('./output');
            res.body.pipe(dest);

            return new Promise((resolve, reject) => {
                dest.on('finish', resolve)
                dest.on('error', reject)
            })
        })
}

const produceCaption = async (file: { buffer: Stream, name: string }, options: {
    model: string;
    response_format: string;
}): Promise<any> => {
    var formdata = new FormData();
    formdata.append("file", file.buffer, file.name);
    formdata.append("model", options.model);
    formdata.append("response_format", options.response_format);
    formdata.append("prompt", "Each part shouldn't be longer than 3 seconds. Each part should have only one sentence and shouldn't have uncompleted sentence.")

    var requestOptions = {
        method: 'POST',
        headers: { "Authorization": `Bearer ${process.env.OPENAI_API_KEY}` },
        body: formdata,
        redirect: 'follow'
    };

    const response = await fetch("https://api.openai.com/v1/audio/transcriptions", requestOptions as any)
    return {
        success: response.status === 200,
        body: response.status === 200 ? await response.text() : await response.json()
    }
}

const inputFileExtension = {
    "flac": ".flac",
    "m4a": ".m4a",
    "mp3": ".mp3",
    "mp4": ".mp4",
    "mpeg": ".mpeg",
    "mpga": ".mpga",
    "oga": ".oga",
    "ogg": ".ogg",
    "wav": ".wav",
    "webm": ".webm"
}

await downloadFile(input.file);
const fileReadStream = fs.createReadStream(`./output`)
const captionResponse = await produceCaption({
    buffer: fileReadStream, name: 'output' + inputFileExtension[input.inputFormat]
}, {
    model: 'whisper-1', response_format: input.outputFormat
});
fs.unlinkSync(`./output`);

await Actor.pushData({
    audioUrl: input.file,
    inputFormat: input.inputFormat,
    outputFormat: input.outputFormat,
    outputText: captionResponse.body

});
await Actor.exit();

await Actor.exit();

src/main.ts

build/types/index.js

import { ProxyConfigurationOptions } from "apify";

export type IInput = {
    file: string;
    inputFormat: "flac" | "m4a" | "mp3" | "mp4" | "mpeg" | "mpga" | "oga" | "ogg" | "wav" | "webm",
    outputFormat: "json" | "text" | "srt" | "verbose_json" | "vtt";
    proxyConfiguration?: ProxyConfigurationOptions
}

Transcribe Audio

Speech Text Actor

Use Cases

Input

Output

Where next?

Format	Description
flac	FLAC audio file
m4a	M4A audio file
mp3	MP3 audio file
mp4	MP4 audio file
mpeg	MPEG audio file
mpga	MPGA audio file
oga	OGA audio file
ogg	OGG audio file
wav	WAV audio file
webm	WEBM audio file

Format	Description
json	JSON format
text	Plain text format
srt	SRT subtitle format
verbose_json	Verbose JSON format
vtt	VTT subtitle format