Abena AI Node.js SDK

The official enterprise-grade Node.js SDK for integrating Abena AI's powerful services into your applications. Access high-quality Text-to-Speech (TTS), accurate Automatic Speech Recognition (ASR), and Neural Machine Translation with a simple and intuitive async/await API.

Features

• High-Quality TTS: Synthesize natural-sounding speech using intuitive voice names.
• Accurate ASR: Transcribe audio from files, Buffers, or Streams.
• Neural Translation: Translate text between hundreds of user-friendly language codes.
• Robust & Secure: Built with production-grade validation and security.
• Detailed Error Handling: Clear, specific exceptions for easy debugging in your application.
• TypeScript Support: Fully typed for a modern development experience.

Installation

The package requires Node.js v14 or higher.

npm install @abena-ai-team/abenasdk

Quick Start

const { AbenasClient, FileError } = require('@abena-ai-team/abenasdk');
const fs = require('fs');

// Best practice: store your API key in an environment variable
// export ABENA_API_KEY="sk_your_api_key_here"
const client = new AbenasClient({
    apiKey: process.env.ABENA_API_KEY
});

async function main() {
    try {
        // Text-to-Speech (TTS)
        const audioBuffer = await client.tts.synthesize(
            'Hello from the Abena AI software development kit.',
            'akua'
        );
        fs.writeFileSync('speech_output.wav', audioBuffer);
        console.log("Audio saved to speech_output.wav");

        // Automatic Speech Recognition (ASR)
        const transcription = await client.asr.transcribe('speech_output.wav', 'en');
        console.log(`Transcription: ${transcription.text}`);

        // Translation
        const translation = await client.translator.translate({
            text: 'Hello from Abena AI!',
            source_language: 'en',
            target_language: 'fr'
        });
        console.log(`Translation: ${translation.translation}`);

    } catch (error) {
        console.error(`An error occurred: [${error.name}] ${error.message}`);
        if (error instanceof FileError) {
            console.error("Please ensure the input file exists.");
        }
    }
}

main();

API Reference

Client Initialization

The client can be configured for production or local development environments.

const { AbenasClient } = require('@abena-ai-team/abenasdk');

// For Production (default)
const client = new AbenasClient({ apiKey: 'YOUR_PROD_KEY' });

Text-to-Speech (client.tts)

synthesize(text, voice, outputFile)

Converts text to speech.

• text (string): The text to synthesize. Max 500 characters.
• voice (string): The voice name to use (e.g., 'akua', 'abena').
• outputFile (string, optional): Path to save the .wav file. If omitted, a Buffer is returned.

// Get audio data as a Buffer
const audioBuffer = await client.tts.synthesize('Hello world', 'akua');

// Or save directly to a file
await client.tts.synthesize('Hello world', 'abena', 'output.wav');

Speech Recognition (client.asr)

transcribe(audioSource, language)

Transcribes an audio file to text.

• audioSource (string | Buffer | Stream): The audio to transcribe. Can be a file path, a Buffer, or a Readable stream.
• language (string): The language of the audio (e.g., 'en').

// From a file path
const result = await client.asr.transcribe('./audio.wav', 'en');
console.log(result.text);

// From a Buffer
const audioBuffer = fs.readFileSync('./audio.wav');
const resultFromBuffer = await client.asr.transcribe(audioBuffer, 'en');

Translation (client.translator)

translate({ text, source_language, target_language })

Translates text between languages.

• text (string): The text to translate. Max 1000 characters.
• source_language (string): The source language code (e.g., 'en').
• target_language (string): The target language code (e.g., 'fr').

const result = await client.translator.translate({
    text: 'Welcome to the future of AI.',
    source_language: 'en',
    target_language: 'twi'
});
console.log(result.translation);

Error Handling

The SDK throws specific, custom errors to allow for robust error handling.

const { AbenasClient, APIError, ValidationError, FileError } = require('@abena-ai-team/abenasdk');
const client = new AbenasClient({ apiKey: '...' });

try {
    await client.asr.transcribe('non_existent_file.wav');
} catch (error) {
    if (error instanceof FileError) {
        console.error(`File Error: ${error.message}`);
    } else if (error instanceof ValidationError) {
        console.error(`Validation Error: ${error.message}`);
    } else if (error instanceof APIError) {
        console.error(`API Error: [Status ${error.status}] ${error.message}`);
        // The raw server response data is available if needed
        console.error('Server Response:', error.data);
    } else {
        console.error(`An unexpected error occurred: ${error.message}`);
    }
}

Documentation & Support

• Full API Docs: https://abena.mobobi.com/playground/sdk/docs/ (Example URL)
• cURL Examples: See the examples/curl/ directory.
• Report an Issue: Please contact [email protected] for any bugs or feature requests.

© 2025 Mobobi LLC. All rights reserved.