I am happy with the results in "Speech Studio" for a sample wav file. How do I scale this up to longer files?

John Woolley 0 Reputation points
2024-04-19T21:50:59.1133333+00:00

I have run a 1-minute wav file through the Speech Studio sample process and am pleased with the result. I can't figure out how to move forward in the system to process larger speech files. One branch seems to take me into a training setting where I learn how to write--perhaps more plausibly to copy--code for several steps. Another branch seems to take me to some place where I am supposed to specify and name "resources." But the form does not actually accept any of the names I make up for resources.

I'm not a coder.

How can it be that there is no intermediate option like the Speech Studio, that lets me generate text from longer sound files??? This is not about $$$; I have a pay-as-you-go account established.The "assistant" answer is not really helpful either because it sends me through an apparently endless number of screens about resources, network security, azure remote desktop software, etc. Jeez, I'm just trying to get text.

Azure AI Speech
Azure AI Speech
An Azure service that integrates speech processing into apps and services.
1,441 questions
{count} votes

2 answers

Sort by: Most helpful
  1. Q&A Assist
    2024-04-19T21:50:59.1333333+00:00
    AI generated content. This question contains an answer created with AI from Q&A Assist. Learn more

    To process longer speech files in Speech Studio, you can use the batch transcription feature. This feature allows you to transcribe multiple audio files at once, up to a total of 10 hours of audio per batch. To use this feature, you will need to upload your audio files to Azure Blob Storage and provide the storage account information in Speech Studio. You can then select the batch transcription option and configure the settings for your transcription job. Once the job is complete, you can download the transcriptions in a variety of formats.

    Here are the steps to use batch transcription in Speech Studio:

    1. Upload your audio files to Azure Blob Storage.
    2. In Speech Studio, select the "Batch transcription" option from the left-hand menu.
    3. Provide the storage account information for your audio files.
    4. Configure the settings for your transcription job, including language, model, and output format.
    5. Start the transcription job.
    6. Once the job is complete, download the transcriptions in the desired format.

    Note that batch transcription is a paid feature and requires a Speech service subscription. For more information on batch transcription and pricing, see the official documentation.


    References:

    0 comments No comments

  2. YutongTie-MSFT 46,996 Reputation points
    2024-04-28T00:03:18.5633333+00:00

    @John Woolley Thanks for your response, I understand you're looking for an easy way to transcribe larger audio files using the Azure Speech Studio without having to write code or configure additional resources.

    While the Speech Studio provides a user-friendly interface for testing and getting transcriptions of smaller audio files, for larger files you will need to use the Azure Speech Services, which does require some coding and configuration.

    Like the batch transcription API, which you mentioned you may want to chunk them into smaller audio, is supported in Speech to Text, please refer to the document here -

    https://learn.microsoft.com/en-us/azure/ai-services/speech-service/batch-transcription

    For a low or no-code solution, you can use the Batch Speech to text Connector in Power Platform applications such as Power Automate, Power Apps, and Logic Apps. See the Power automate batch transcription guide to get started.

    Please take a look and see how it works on your case. I hope this helps!

    Regards,

    Yutong

    -Please kindly accept the answer if you feel helpful to support the community, thanks a lot.

    0 comments No comments