Azure-docs: Is JavaScript officially supported?

[compulim]

The JavaScript SDK can be found at https://www.npmjs.com/package/microsoft-speech-browser-sdk and published by Microsoft. Is it officially supported? If yes, please add it to the list.

---

Document Details

⚠ Do not edit this section. It is required for docs.microsoft.com ➟ GitHub issue linking.

• ID: efae7ff6-5557-221f-7539-5d2d1883127d
• Version Independent ID: 4b868e15-0248-635c-aa2f-d591bbc77d65
• Content: About the Cognitive Services Speech SDK - Microsoft Cognitive Services
• Content Source: articles/cognitive-services/Speech-Service/speech-sdk.md
• Service: cognitive-services
• GitHub Login: @v-jerkin
• Microsoft Alias: v-jerkin

[MohitGargMSFT]

@compulim Thanks for reporting! I have assigned the issue to the content author to investigate and update as appropriate.

[v-jerkin]

One of the developers should take this one. I've let them know.

[zhouwangzw]

The JavaScript SDK is still supported. However it has a different API. We will update the JavaScript SDK to have API parity with the current speech SDK.

[sfamini]

Is there sample code for continuous recognition with javascript?
The github sample code repository only has samples for one-shot recognition. Trying to replicate the continuous functionality available in C# samples results in errors from ts files (in Chrome debugger).

[compulim]

@zhouwangzw do you know when will it be updated? Is the updated version means the ZIP file in this article?

[fmegen]

We are currently preparing sample code for the APIs.
basically, you can follow the steps to create a speech recognizer from the quickstart,

            reco = new SpeechSDK.SpeechRecognizer(speechConfig, audioConfig);

then wire up handlers:

            reco.recognizing = function (s, e) {
                window.console.log(e);
            };

            // The event recognized signals that a final recognition result is received.
            // This is the final event that a phrase has been recognized.
            // For continuous recognition, you will get one recognized event for each phrase recognized.
            reco.recognized = function (s, e) {
                window.console.log(e);
            };

finally, start the recognizer:
// Starts recognition
reco.startContinuousRecognitionAsync();

hope that helps for the moment.

[compulim]

Will try it out.

But a SDK released in a ZIP file and not on NPM is not consumable in our production environment.

[sfamini]

@fmegen: That helped a lot, thank you very much!

I was doing it this way, which was giving me no results:

...
          recognizer.startContinuousRecognitionAsync(
                    function (result) {
                        phraseDiv.innerHTML += result.text;
                        window.console.log(result);
                    },
                    function (err) {
                        phraseDiv.innerHTML += err;
                        window.console.log(err);
                    }) ;
...

[fmegen]

@sfamini: The callbacks to the startContinousRecognitionAsync only report the success/failure of the start operation itself and do not report any results.
you need to register explicitly for the events in this scenario.

kind regards,
friedel

[fmegen]

We will update the demo repository soon. Please checkout
https://github.com/Azure-Samples/cognitive-services-speech-sdk

[npacucci]

@compulim i'm agree with you.
This newest SDK will be released also on NPM?

[fmegen]

We are working on NPM but there is no confirmed date yet.

[tchristiani]

please-close