I need some guidance. My Azure function (written in node.js) will convert some random text to speech and then upload the speech output to a Blob. I will like to do so without using an intermediate local file. BlockBLobClient.upload method requires a Blob, string, ArrayBuffer, ArrayBufferView or a function which returns a new Readable stream, and also the content length. I am not able to get these from the RequestPromise object returned by call to TTS (As of now I am using request-promise to call TTS). Any suggestions will be really appreciated.
Thank you
Adding a code sample that can be tested as "node TTSSample.js" Sample code is based on
Azure Blob stream related code shared at https://github.com/Azure-Samples/azure-sdk-for-js-storage-blob-stream-nodejs/blob/master/v12/routes/index.js
Azure Text to speech sample code at https://github.com/Azure-Samples/Cognitive-Speech-TTS/tree/master/Samples-Http/NodeJS
Replace appropriate keys and parameters in the enclosed settings.js
I am using node.js runtime v12.18.3
Input text and the output blob name are hard coded in this code sample.
// TTSSample.js // TTSSample.js // Converts given text to Audio and uploads to Azure Blob storage // Ultimately to be used in an Azure function // As of now input text as well as the target BLobStorage object are hard coded. // To install dependencies, run: npm install const xmlbuilder = require('xmlbuilder'); // request-promise has a dependency on request const rp = require('request-promise'); const fs = require('fs'); const readline = require('readline-sync'); const settings = require('./settings.js'); const multer = require('multer'); const inMemoryStorage = multer.memoryStorage(); const uploadStrategy = multer({ storage: inMemoryStorage }).single('image'); const getStream = require('into-stream'); const ONE_MEGABYTE = 1024 * 1024; const uploadOptions = { bufferSize: 4 * ONE_MEGABYTE, maxBuffers: 20 }; const ONE_MINUTE = 60 * 1000; // Blob Storage const { BlobServiceClient } = require('@azure/storage-blob'); // Gets an access token. function getAccessToken(subscriptionKey) { let options = { method: 'POST', uri: settings.issueTokenUri, headers: { 'Ocp-Apim-Subscription-Key': subscriptionKey } } return rp(options); } // Converts text to speech using the input from readline. function textToSpeech_rp(accessToken, text) { // Create the SSML request. let xml_body = xmlbuilder.create('speak') .att('version', '1.0') .att('xml:lang', 'en-us') .ele('voice') .att('xml:lang', 'en-us') .att('name', 'en-US-Guy24kRUS') // Short name for 'Microsoft Server Speech Text to Speech Voice (en-US, Guy24KRUS)' .txt(text) .end(); // Convert the XML into a string to send in the TTS request. let body = xml_body.toString(); //console.log("xml string is done"); let options = { method: 'POST', baseUrl: settings.cognitiveUri, url: 'cognitiveservices/v1', headers: { 'Authorization': 'Bearer ' + accessToken, 'cache-control': 'no-cache', 'User-Agent': settings.cognitiveResource, 'X-Microsoft-OutputFormat': 'riff-24khz-16bit-mono-pcm', 'Content-Type': 'application/ssml+xml' }, body: body } console.log(options); let request = rp(options) .on('response', async (response) => { if (response.statusCode === 200) { console.log("Inside response"); const stream = getStream(response); //request.pipe(fs.createWriteStream('TTSOutput.wav')); const AZURE_STORAGE_CONNECTION_STRING = settings.storageConnectionString; const storageAccount = settings.storageAccount; const storageKey = settings.storageKey; const containerName = settings.audioContainer; // Create the BlobServiceClient object which will be used to create a container client const blobServiceClient = BlobServiceClient.fromConnectionString(AZURE_STORAGE_CONNECTION_STRING); // Get a reference to a container const containerClient = blobServiceClient.getContainerClient(containerName); // Create a unique name for the blob //const blobName = id + '.mp3'; const blobName = 'audio1.mp3'; // Get a block blob client const blockBlobClient = containerClient.getBlockBlobClient(blobName); //blockBlobClient.upload(buffer,61000); //blockBlobClient.upload(request.body); try { await blockBlobClient.uploadStream(stream, uploadOptions.bufferSize, uploadOptions.maxBuffers, { blobHTTPHeaders: { blobContentType: "audio/mpeg3" } }); //res.render('success', { message: 'File uploaded to Azure Blob storage.' }); console.log('Success'); } catch (err) { //res.render('error', { message: err.message }); console.log("Failure", err.stack); } console.log("I am done"); } }); return request; }; // Use async and await to get the token before attempting // to convert text to speech. async function main() { const subscriptionKey = settings.subscriptionKey; //process.env.SPEECH_SERVICE_KEY; if (!subscriptionKey) { throw new Error('Environment variable for your subscription key is not set.') }; // Prompts the user to input text. //const text = readline.question('What would you like to convert to speech? '); const text = "Hello there"; try { const accessToken = await getAccessToken(subscriptionKey); //console.log("Access Token is done"); await textToSpeech_rp(accessToken, text); } catch (err) { console.log(`Something went wrong: ${err}`); console.log(err.stack); } } // Run the application main() //============ Settings.js (function() { "use strict"; module.exports = { // Replace with your own subscription key, service region (e.g., "westus"), // and recognition language. subscriptionKey: "CognitiveServiceSubsctiptionKey", serviceRegion: "eastus", // e.g., "westus" language: "en-US", issueTokenUri: 'https://eastus.api.cognitive.microsoft.com/sts/v1.0/issuetoken', cognitiveUri: 'https://eastus.tts.speech.microsoft.com/', cognitiveResource: 'CognitiveResourceName', // Storage storageAccount: "StorageAccount", storageKey: "sN/StorageKey==", storageConnectionString: "StorageConnectionString", audioContainer: "AudioContainer", }; }());
