Australia
Asked — Edited

Inconsistent Speed Speech Settings After Installing Cepstral Voices

After installing some Ceptral voices and adjusting some Windows settings, speech synthesis through ARC is suddenly much faster than before. I used another laptop without Cepstral voices installed to test how it originally sounded. After ensuring that the Text-to-speech and Narrator settings were the same for the computer and laptop (voice speed settings set to 10), the "Medium" Speech Settings Rate on the laptop is the same speed as "ExtraSlow" on the computer.

User-inserted image

User-inserted image

Both the computer and the laptop run Windows 10.

I'm fairly certain this problem is related to the windows settings but I've tried changing the Text-to-speech and Narrator settings but these don't fix it.

The speed is still accelerated regardless of the voice chosen, be it MS Zira Desktop or MS David Desktop. Cepstral voices also sound accelerated but I can't really compare since they're not installed on the laptop. From what I remember, their speed was normal when first installed.

Does anyone have any ideas on how to reset the voice speed for the computer back to normal?


Related Hardware JD Humanoid

ARC Pro

Upgrade to ARC Pro

Get access to the latest features and updates before they're released. You'll have everything that's needed to unleash your robot's potential!

Australia
#2  

Yep, the voice speed settings are at 10 for both Text-to-speech and Narrator as well as for both the laptop and the computer. These settings don't seem to affect ARC. Even when the speeds on Windows are changed, ARC's speeds for each of its "ExtraSlow", "Slow", "Medium", "Fast" and "ExtraFast" remain unchanged.

PRO
USA
#3  

Those settings are related to universal applications they are not relevant to desktop win32 applications e.g. ARC.

please read carefully the link i mentioned before.

the relevant settings are available through the control panel.

did you read the link ?

Australia
#4  

Ah, right. I skimmed over your link and missed the part where it said that Windows 10 has two places for Text to Speech settings. The problem is solved now. Thank you!

#5  

can use it for speech recognition?

PRO
USA
#6  

@joinny can you be more specific about the WHAT ?

#7  

ptp I am struggling to convert the speech recognition language into Korean, I tried everything but it doesn't seem to work well and it still displays the initial default language en-us english. i hope you can help me!

PRO
USA
#8  

The original question is related to Text to Speech. Cepstral https://www.cepstral.com/ asfaik only sells custom voices for Text To Speech. Windows 10 supports Korean Text to Speech, when you install Korean language support with voice add on automatically adds a Microsoft Heami Desktop voice and can be used with ARC Text To Speech control.

Speech Recognition or Speech to Text requires a language recognizer, and Windows 10 does not install one. So it's not a configuration problem neither ARC can solve the problem.

The solution for your problem is to use a Cloud service e.g. Google, Microsoft Azure and IBM Watson all of them (requires additional confirmation) support Korean Speech To Text. A cloud service requires an account with service subscription. ARC has plugins for Microsoft, Google, IBM.