This article is contributed. See the original author and article here.
Custom Neural Voice, a feature of Azure Cognitive Services for Speech, is a great way to create a highly natural synthetic voice that sounds almost identical to your voice actor. This synthetic voice can then be used in a variety of different scenarios including audiobooks, language learning, reading news content and many more. Since its launch, Custom Neural Voice has empowered organizations such as AT&T, Duolingo, Progressive, and Swisscom to develop branded speech solutions that delight users. (For more details, read the Innovation Stories blog).
Microsoft has made it simple to train a professional custom neural voice by using a small set of recordings from the target voice (from 300 to 2,000 sentences or short phrases, which is about 30 minutes to 3 hours of speech data). However, the studio recording process takes time and many customers are looking for a way to try the voice customization capability more easily.
Today we are glad to introduce Custom Neural Voice Lite, a new feature in public preview, which enables users to clone their voice by recording just 5 minutes of speech data. This new feature makes it extremely easy for customers to create a synthetic voice that sounds natural.
Custom Neural Voice Lite
Custom Neural Voice (CNV) now supports two project types, Pro and Lite. The Pro version is best for professional scenarios like brand and character voices for chat bots, or audio content reading. In this blog, we provide detailed instructions on how to create a professional custom neural voice. The new Lite version is best for producing quick demos or creating personal voice clones.
Due to the sensitivity of the technology, we have limited the access and use of Custom Neural Voice. However, every customer with a valid Azure Speech resource can create CNV Lite voices by recording their own voice for evaluation purposes. After creating a CNV Lite voice, the customer must submit an application with their use case to gain full access to the Custom Neural Voice capability before they can use the voice for business scenarios.
The following table summarizes the key differences between the CNV Pro and CNV Lite project types.
Items
Lite (Preview)
Pro
Target scenarios
Demonstration or evaluation
Professional scenarios like brand and character voices for chat bots, or audio content reading
Training data
Record online from your own computer using Speech Studio
Bring your own data. Recording in a professional studio is recommended.
Scripts for recording
Provided in Speech Studio
Use your own scripts that match the use case scenario. Microsoft provides example scripts for reference.
Required data size
20-50 utterances
300-2,000 utterances
Training time
Less than 1 compute hour
Approximately 20-40 compute hours
Voice quality
Moderate quality
High quality
Availability
Anyone can record samples online and train a model for demo and evaluation purpose. Full access to Custom Neural Voice is required if you want to deploy the CNV Lite model for business use.
Data upload is not restricted, but you can only train and deploy a CNV Pro model after access is approved. CNV Pro access is limited based on eligibility and usage criteria. Request access on the intake form.
Pricing
Per unit prices apply equally for both the CNV Lite and CNV Pro projects. Check the pricing details here.
Per unit prices apply equally for both the CNV Lite and CNV Pro projects. Check the pricing details here.
To get an idea of how a Lite voice sounds like, check the samples below.
Language
Human recording
TTS (CNV Lite)
English
Chinese
How it works
A Speech service resource is required before you can create a Custom Neural Voice project. If you do not have a Speech resource in Azure, follow these instructions to create one. Make sure you select one of these regions for your resource – East US, Southeast Asia, or UK South where Custom Neural Voice training is supported. Select S0 for the pricing tier. Free tiers are not available for Custom Neural Voice.
Creating a Speech resource
To build your CNV Lite voice, go to Speech Studio. Log in with the right Speech resource selected. Then click on the ‘Custom Voice’ tile and select to create a Custom Neural Voice Lite project. CNV Lite now supports English and Chinese (Mandarin).
Creating a CNV Lite project
Once the project is successfully created, you can start to build your voice. Before you move forward, make sure you read and understand the Voice Talent Terms of Use, and provide your agreement for Microsoft Speech Studio to collect your voice data (at this step, for evaluation purpose). To protect each user’s voice identity, the Lite project will be removed within 90 days if your company does not have its business use case approved by Microsoft (check the limited access policy), or the voice talent whose data is used for training does not provide explicit agreement for using his/her voice to generate synthetic speech outside of the evaluation purpose (check the voice talent disclosure requirement).
Once you have accepted the terms of use, you can start to record your voice samples. Read the recording instructions carefully. The quality of your recording data is critical to the training output. Check your environmental noise and do not record if noises are detected.
Noise check before voice recording
Tips for recording:
Increase the clarity of your samples by using a high-quality microphone. Speak about 8 inches away from the microphone to avoid mouth noises.
Relax and speak naturally. Allow yourself to express emotions as you read the sentences.
To keep a consistent energy level, record all sentences in one session.
Pronounce each word correctly and speak clearly. After recording each sample, check its quality metric before continuing to the next one.
Although you can create a model with just 20 samples, it’s recommended that you record up to 50 to get better quality.
After each sample is recorded, double check the audio quality before you click to record the next. Several metrics are provided to help you review the quality, enabled with the pronunciation assessment technology.
As shown in the screenshot below, mispronunciations are automatically detected on each audio. It’s recommended that you make sure your recorded audio is green with accepted quality.
“Clearness” indicates the speech signal against the noise. You get a higher clearness score if the noise level is lower.
“Pronunciation” shows the accuracy of your pronunciation at the sentence level. You should make sure you pronounce each word correctly with no omission or insertion.
“Volume” of your voice in the recording should be kept stable. Don’t speak too far or too close to your mic. An audio that’s too loud or too low volume is not acceptable.
Recording voice samples with quality check
After you have recorded at least 20 samples, and checked the quality is all good, you can click the ‘Train model’ button at the bottom of the page to start your voice training. It’s estimated that each training takes about 40 minutes. Check the pricing page to get an idea of the cost before you hit ‘Create’.
Once the model is successfully created, you can listen to the sample output for demo and evaluation purpose.
Sample output of a CNV Lite voice model
To deploy your voice model and use it in your applications, you must get full access to Custom Neural Voice and explicit consent from your voice talent. You can submit a request form here. For guidance on applying for Custom Neural Voice, you can watch this short video. With the full access approved, you can get your CNV Lite voice integrated with your apps, or move to create a CNV Pro project with professional studio recordings for an even more natural voice. Check this blog for the instructions to create a high-quality professional voice.
Learn more
We are excited about the future of Neural TTS with human like, diverse and delightful quality under the high-level architecture of XYZ-Code AI framework. Our technology advancements are also guided by Microsoft’s Responsible AI process, and our principles of fairness, inclusiveness, reliability & safety, transparency, privacy & security, and accountability. We put these ethical standards into practice through the Office of Responsible AI (ORA), which sets our rules and governance processes, the AI, Ethics, and Effects in Engineering and Research (Aether) Committee, which advises our leadership on the challenges and opportunities presented by AI innovations, and Responsible AI Strategy in Engineering (RAISE), a team that enables the implementation of Microsoft responsible AI rules across engineering groups.
This article is contributed. See the original author and article here.
CISA and the Department of Energy (DOE) are aware of threat actors gaining access to a variety of internet-connected uninterruptable power supply (UPS) devices, often through unchanged default usernames and passwords. Organizations can mitigate attacks against their UPS devices, which provide emergency power in a variety of applications when normal power sources are lost, by removing management interfaces from the internet.
Organizations can mitigate attacks against UPS devices by immediately removing management interfaces from the internet. Review CISA and DOE’s guidance on mitigating attacks against UPS devices for additional mitigations and information.
This article is contributed. See the original author and article here.
Today we are announcing the general availability of on demand capacity reservations for Azure Virtual Machines. This new feature allows you to deploy and manage the compute capacity required to run Azure VMs separately from the VMs themselves.
On-demand capacity reservations create new options for you to better manage your compute capacity needs. During our discussions with Azure customers, they often identify the following scenarios:
Business-critical applications — use on-demand capacity reservations to protect their capacity, for example when taking these VMs offline to perform updates.
Disaster recovery (DR) — set aside compute capacity to ensure a seamless recovery in the event of a natural disaster. The compute capacity can be repurposed to run other workloads whenever DR is not in effect. The VM maintenance can be handled by keeping core images up to date without the need to deploy or maintain VMs outside of DR testing.
Special events—claiming capacity ahead of time provides assurance that your business can handle the extra demand.
On demand capacity reservations come with a capacity SLA. Unused reserved capacity and Virtual Machines using reserved capacity are both eligible for Azure Reserved Virtual Machine Instance term discounts. However, Reserved VM Instances and on-demand capacity reservations are very different and offer two separate sets of benefits captured in the table below:
Additional Azure services are integrating with on-demand capacity reservations. The Azure Site Recovery service has public preview support for attaching reserved capacity for DR purposes.
Existing and future capabilities
The number of Azure Virtual Machines series and sizes supported has increased since the start of the public preview.
Specifically, the following Azure Virtual Machines series are now supported:
Burstable (B-series VMs)
General-purpose VMs (Av2, Dv2-series VMs and newer generations including Dv5 and Dasv5-series VMs, including support for both Intel and AMD-based VM sizes)
Memory-optimized VMs (Ev3-series VMs and newer generations including Ev5 and Easv5-series VMs; both Intel and AMD-based VM sizes)
Compute-optimized VMs (Fv1-series VMs and newer generations)
Microsoft is finalizing support for additional Azure Virtual Machines series. At this time, we expect previews in Q2/Q3 2022, followed by general availability, for the following VM series:
Storage optimized VMs (L-series)
Additional memory-optimized VMs (M-series)
GPU-enabled VMs (NC-series and NV-series)
Azure also offers VM series using InfiniBand networking: ND, HB, and HC VM series. These series will be supported in the future.
The VM deployment methods supported with on demand capacity reservations during public preview are now supported with general availability: single VMs and Virtual Machine Scale Sets using uniform orchestration mode. Starting today, on-demand capacity reservations for Virtual Machine Scale Sets flexible orchestration mode is in public preview. Please read the documentation to learn more.
This article is contributed. See the original author and article here.
Google has released Chrome version 99.0.4844.84 for Windows, Mac, and Linux. This version addresses a vulnerability that an attacker could exploit to take control of an affected system.
CISA encourages users and administrators to review the Chrome Release Note and apply the necessary updates.
This article is contributed. See the original author and article here.
From time to time we have users who open the same question multiple times.
The Issue – direct result of duplicate threads:
(1) Supporters – People that come to help might waste their time on something that was answered in the other thread.
(2) Users – people that search for the same issue might find the thread which has less information and miss the information in the other thread.
(3) original poster (OP) – the person who asked the question will probably not remember to follow all the threads which he opened.
Supporters and Users will waste time on responds which no one will read.
OP might miss the best answer/discussion while following a different thread.
In short, this is a lose-lose case where everyone lose!
Best practice for Moderators
In each interface the features which are built-in are different so these option might not fit all forums interfaces. With that said, as I want to focus mainly on the QnA forums, I will provide my insights according to the QnA features exists at this time.
The list is sorted from the best option for first time case – note that a user which continue such action is a different story. I the OP was inform on the issue and was asked to avoid such cases, then his behavior can be considered as abusive and should be treated accordingly.
In any case, the options which prevent such issue in advance, will be best!
Clear Policy document
In any communication interface, there must be a clear policy document which we can point the OP to. A link to the policy document must be presented in a way that no one will miss it and that everyone can view it from any page where the OP post his question. Usually it is recommended to add a clear link at the top of the page.
An intellisense feature
An intellisense feature provides information while the user type the content. It is well common in code editors in order to provide code completion, parameter info, quick info, and member lists. In the scope of the the discussion an intellisense feature will provide the user a list of previous thread according to the information he is typing.
Note! This feature existing at Microsoft QnA forums.
Insights!
Do not assume that the OP saw the forum policy or noticed the content of the intellisense information! Most people are focusing on their needs and ignore “background noises”. It may not be the most positive behavior but it is definitely a natural behavior.
Contact the OP in private
How?
If the system include an option to send internal private message then this is your best action!
The user email should be visible to other users, but if this information is available to you and there is no internal messaging feature, then send the user an email.
If there is no build-in messaging feature, then check the user profile and signature for links to his social media network (Facebook, Twitter, linkedin and so on) and send him a private message.
What?
Inform the user about the forum policy, add a link to the official policy document if exists and ask him to avoid such cases in the future. Don’t forget to send him links to all the duplicate threads which you found. Ask the user to select one of the threads, in all other thread add a link to the active thread and close the rest of the threads as duplicated.
Note! This feature not available to community moderators at Microsoft QnA forums.
Contact the OP in public
Add a response to the problematic message(s).
In most social media network a user which is not connected to you will not see your message. In this case you can add a response to one of the user last discussions. Remember that this is a public message and you should be extremally polite! Do not use this option if not must.
Provide the same information as in the case of private message
Note!
Taking actions behind the scenes has no value for the future behavior of the user and might lead the user to re-do the same, as he was not inform about an issue – this only raise the issue as it lead for more duplicate threads (actions behind the scene are for example deleting message, reporting message, and so on).
If you chose to take an action behind the scenes then you should also try to inform the OP about the taken actions!
Merge duplicate threads
By merging the duplicate threads you can ensure that no information is lost. Users who go to each of the links to the separate threads come to the same place – the merged thread. With that said, this action makes it a little difficult to orient in a discussion because one discussion combines responses from several discussions and the order of the messages can be confusing.
Note! This feature NOT existing at Microsoft QnA forums. It is exist in the MSDN forums.
Lock the duplicate thread(s) & comments
By locking the duplicate threads you can ensure that no information is lost. Users can still navigate to each of the links and watch the separate discussion. It is HIGHLY important to add a message that explain why this thread was locked and provide a link to the active thread.
In the QnA system locking the thread is done by closing the thread. It is important to know that close the thread does not prevent people from adding comments! You should go over each of the messages in the thread and lock the comment!
Note! This feature existing at Microsoft QnA forums.
Redirect the duplicate thread(s) to the one selected for continuing the discussion.
Redirecting users from a thread to another thread has a very problematic side effect: The original link still exists but it cannot be used to get the original message. This makes it harder to report the issue since the link redirect to an active thread and it make it impossible to inform the user about the issue and show him the duplicated thread.
Do not! (on the first case)
Do not treat the user as abuser! It can be a mistake.
Do not delete a message without informing the user. He might not find the message and create another new message.
Do not report the message without informing the user at the same time, since he cannot see the report and know about the issue.
Do you want to add your insights? You can add comments to this post or contact me in private if you want me to update the post, and in the meantime just remember to have fun and continue helping others.
Recent Comments