Practical Guide to Azure Custom Neural Voice: Essential Tips for Success

sound editing software

There are several possible solutions, such as Adobe Audition or Audacity. I recommend using Audacity. Not only because Adobe Audition is paid, but also because Audacity’s limited features are ideal for our needs. We just need to select the speech, export it, and cut it. Minimalism is the key to success. Audacity also makes it easy to navigate the track and minimizes the unnecessary toolbox.

The File menu in Audacity provides commands to create, open, and save projects, and import and export audio files. For example, the Export function is not assigned by default, so you can easily create a shortcut to export a selection. This speeds up the process considerably. In my experience using both Adobe Audition and Audacity, I was able to complete the same amount of work in two days using Audacity, compared to four days using Adobe Audition.

price

Here are my project details:

Pricing may vary depending on engine version and number of training hours, but you will at least get a sample.

Intake form

You probably know that access is granted only after you complete the Intake Form and that decisions are made based on eligibility and usage criteria. Before providing any project information, please refer to the following: Microsoft’s Responsible AI StandardsThis will allow you to tailor your description and scenario accordingly.

Prepare your audio

The process is very simple. Create a notepad with all the utterances and their IDs. Select the utterances one by one, export them, save them with their IDs, and then delete them from the notepad. Define the optimal size in advance and do not zoom in or out while working. You will become familiar with the timeline size and will be able to add the 100-200 milliseconds of silence you need more easily.