![]() Gilbert Segura is the CTO at Global eLearning.Amazon Transcribe makes it easy for developers to add speech-to-text capability to their applications. If you want to learn more about TTS, voice talent, localization, and translation services, Global eLearning is widely considered to be a leader in TTS, especially for the learning and development industry. Sometimes having a voice, while not perfect, is better than none while, at other times, it’s worth the additional cost and effort for a studio recording. We believe your stakeholders and end customers should help guide what’s appropriate and best for your specific application. There remains a lot of variability but overall the non-neural voices are not as high quality as others since they rely on parametric models–and are mostly on-par with Google’s non-WaveNet offerings. Overall, they are acceptable for the cases where no alternative exists. Quality: The 5 Neural TTS offerings are almost a start–but they don’t really have the breadth of offerings as the other vendors.This however is “gated technology” meaning it’s potential for abuse and other ethical concerns means that you have to invest heavily and get approval from Microsoft for your use. This opens the door to customization and allows you to take a regular TTS voice and truly model one for yourself. Microsoft’s Special Add-ons – The speech service is able to be trained with additional data for 9 major languages.With 81 voices though, it’s somewhat shallower than the others meaning sometimes you only have a single voice to choose from. The service lists support for 49 languages and 81 voices. It is worth noting that Microsoft offers some languages that are not available on AWS or Google–so in terms of coverage this is the broader list. Coverage – Microsoft Azure Text to Speech is probably the oldest entry for TTS and they even have a Windows-based SAPI for screen readers and other software in Windows.So, for the basic user, this is probably not the best option. To get it to work, they do provide some sample code but it assumes you know something about C# or Python. As a large developer-run organization, it makes sense that the TTS options on their cloud platform share tame terminology and usage as their sophisticated tools for developers–not end-users. Namely, you’ll need to get an API and be a developer or have access to one. Ease of Use – The TTS Options offered by Microsoft Azure Text to Speech offer a similar approach to Google when it comes to availability for using their product.So, while it looks similar to Google’s offering, there’s some potential golden nuggets, hidden in the architecture. Basically, they skip the entire process of anything written – just talk in one language and it will talk back in the other. Respectively, these are different approaches to transcription and speech generation. In this platform offering, you can customize and augment the speech service–which is packaged closely with Speech to Text, Text to Speech, and Speech to Speech. TTS Options offered by Microsoft Azure Text to Speech provides a reliable, if somewhat less exciting collection, of voices with a very interesting mix of back-end abilities. Your rating on quality will directly correlate to the original goals you outlined. ![]() But in some cases, such as French, you may have up to 10 options between all 3 providers with multiple genders and accents.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |