Extend Google Cloud Text To Speech Tool From Python Library

[asadCoder]

Checked

• I searched existing ideas and did not find a similar one
• I added a very descriptive title
• I've clearly described the feature request and motivation for it

Feature request

Feature request to introduce the google cloud text-to-speech API to JS library.

Adding this to the JS library would:

• Ensure Consistency: Align functionality across both Python and JS libraries
• Enhanced Capabilities for Voice-Driven Applications: Give developers at least one toolkit option for text-to-speech in the JS library

Motivation

Currently, the Langchain Python library has integrated support for the Google Cloud Text-to-Speech API, providing developers speech synthesis capabilities. However, there is no equivalent support for the JS library, leading to inconsistency across Langchain's ecosystem.

Proposal (If applicable)

We are a team of four, from the University of Toronto, eager to contribute to the open-source community. We intend to create and complete a pull request for this feature by no later than mid-November.

Sample Usage and Output

import { Tool } from 'langchain/tools';
import { TextToSpeechClient } from '@google-cloud/text-to-speech';
import fs from 'fs';

class GoogleCloudTextToSpeechTool extends Tool {
    constructor() {
        super();
        this.client = new TextToSpeechClient(); 
    }

    async _call(input) {
        const request = {
            input: { text: input.text },
            voice: { languageCode: 'en-US', ssmlGender: 'NEUTRAL' },
            audioConfig: { audioEncoding: 'MP3' },
        };

        const [response] = await this.client.synthesizeSpeech(request);
        const outputFile = `./output.mp3`;
        fs.writeFileSync(outputFile, response.audioContent, 'binary'); 

        return outputFile; // Return file path
    }

    name() { return 'google_cloud_texttospeech'; }
}

// Usage Example
const tool = new GoogleCloudTextToSpeechTool();
tool._call({ text: "Hello, welcome to LangChain!" }).then((outputFile) => {
    console.log(`Generated audio file: ${outputFile}`);
});

[asadCoder]

@jacoblee93 What do you think of this? I think it will be useful to maintain consistency between both JS and Python libraries. If so, could we turn this into an issue and start working on it?

[Minjun1Kim]

This is a brilliant suggestion! My project could really benefit from this implementation in JS. Looking forward to it

[danishmohammed]

I'm really looking forward to this idea being implemented!