Microsoft Cognitive Speech SDK supporting multiple language translations

Question

Microsoft Cognitive Speech SDK supporting multiple language translations

DevUser2048 0

Hi,

I'm using Microsoft Cognitive Speech SDK node package to do the speech translation. With our requirement, we wanted to add support translations to multiple target language. I see it's working when we add the target language while creating the instance of SpeechTranslationConfig.

But in our case, the target languages are decided at a later point of time based on who are the various users who join the discussion. So instead of recreating the SpeechTranslationConfig instance everytime, can we add the target language after the instance is created?

const speechTranslationConfig = SpeechTranslationConfig.fromAuthorizationToken(token,region);

speechTranslationConfig.speechRecognitionLanguage = "source language";
speechTranslationConfig.addTargetLanguage('targetLocale1');
speechTranslationConfig.addTargetLanguage('targetLocale2');

const audioConfig = getAudioConfig();
const speechRecognizer = new TranslationRecognizer(speechTranslationConfig, audioConfig);

1 answer

Your answer

Answer 1

DevUser2048 0

Looks like it's possible. We have to add the language to the speechRecognizer instance rathre than the speechTranslationConfig instance.

navba-MSFT 27,550 Reputation points Microsoft Employee Moderator

2023-10-30T06:00:45.6133333+00:00
@DevUser2048 Welcome to Microsoft Q&A Forum, Thank you for posting your query here!

I see that you are using the Microsoft Cognitive Speech SDK node package to do speech translation. You want to add support for translations to multiple target languages, but the target languages are decided at a later point in time based on who joins the discussion. You want to know if you can add the target language after the SpeechTranslationConfig instance is created.

Here is the official sample code Azure-Samples/cognitive-services-speech-sdk/samples/js/node/translation.js for Speech translation.

You can add target languages to an existing SpeechTranslationConfig instance. You can use the addTargetLanguage method to add a new target language to the existing instance.

// Create the translation config with the target language(s). var translationConfig = SpeechTranslationConfig.fromSubscription(subscriptionKey, serviceRegion); translationConfig.addTargetLanguage("de"); // German translationConfig.addTargetLanguage("fr"); // French translationConfig.addTargetLanguage("zh-Hans"); // Simplified Chinese // Create the audio config pointing to our stream and // the speech config specifying the source language. var audioConfig = AudioConfig.fromDefaultMicrophoneInput(); translationConfig.speechRecognitionLanguage = "en-US"; // English // Create the translation recognizer. var recognizer = new TranslationRecognizer(translationConfig, audioConfig); // Start the recognizer and wait for a result. recognizer.recognizeOnceAsync( function (result) { console.log(result); recognizer.close(); recognizer = undefined; }, function (err) { console.trace("err - " + err); recognizer.close(); recognizer = undefined; } );

This example uses single-shot recognition, which recognizes a single utterance. You can also use continuous recognition to control when to stop recognizing. The addTargetLanguage method is used to specify the target languages for translation. In this example, English speech is translated into German, French, and Simplified Chinese.

I see that you have tried adding the language to the speechRecognizer instance rather than the speechTranslationConfig instance. Please let me know if you face any issues. I would happy to help.

Share via

Microsoft Cognitive Speech SDK supporting multiple language translations

1 answer

Your answer