This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(240, 34%, 33%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ERP%3C/text%3E%3C/svg%3E)
I have a question regarding cognitive services, specifically for near real-time speech-to-text and pronunciation. What would be the best approach for streaming the audio directly from a reactjs application to Azure, getting the cognitive service response and at the same time storing the audio in a bucket? The bucket content would have to be later available for a different application.
I am trying to avoid the option of recording the user input first, then separately sending the file to the storage bucket and Azure cognitive.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(22.400000000000002, 11%, 12%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ERM%3C/text%3E%3C/svg%3E)
@Rob P Thanks for the question. Currently Batch transcription is a set of REST API operations that enables you to transcribe a large amount of audio in storage.
Here is the document for STT limits.
You can use the video indexing service utilizes the cognitive speech to text service as a part of its processing. The video index service may do some pre-processing of the audio before sending to cognitive speech service, may do some post processing of the resultant text, and utilize parameters of the cognitive speech service that is specific to the task of video indexing.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(278.4, 1%, 36%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EYM%3C/text%3E%3C/svg%3E)
@Rob P Thanks for the question again. Do you have a chance to check on Ram's response? Please let us know if you have more questions about this issue, we are willing to help.
Regards,
Yutong