Voice technology for the next generation

CopyAMP code

From smart speakers in homes to voice controls in cars, voice technology has been steadily gaining popularity in recent years. The COVID-19 pandemic has increased the popularity of voice technology, meaning about 40 percent of the US population will be using a smart speaker every month by 2020. The acceptance of voice technology will increase in 2020 due to its unrelated appeal, and now that we have reached the vaccination phase, brands have realized that consumers do not want to give up the convenience of advancing voice technology.

CopyAMP code

As people continue to anticipate the simplicity and convenience of voice-powered communication, companies must first think about developing and using a conversational approach to build deeper relationships. for customers. Digital voice technology can help companies bridge the gap between their products and services for their valued consumers.

To come up with a successful digital voice strategy, companies must first understand what a digital voice is and how they can use it to their advantage.
What is digital voice?

Of the 61 percent of adults who already use voice assistants to report repeated use in the future, such as to complete tasks such as buying or adding items to a shopping list, companies have access to these devices. But there is an opportunity to meet people. However, companies can move beyond voice assistants built into smart speakers, creating a unique voice that speaks for the company’s brand, and outsmarting them with the constant sound of the assistant’s voice. Custom digital voices can be used across a variety of devices and touch points to create a seamless experience and enhance brand recognition throughout the consumer journey. Like a network of visual branding, the voice used by a brand should be instantly recognizable, engaging, memorable, and consistent across all devices and platforms. When consumers hear the same voice – regardless of device or platform – they trust the brand more and form an emotional connection, which is important for brands as voice technology continues to evolve.

Companies need to choose the right voice partner to grow digital voice. Ideally, the company chooses a partner who guarantees data privacy, quality assurance, and uses the best technology available so that the company can create the best experience while guaranteeing consumers’ privacy. .
Next generation digital voice enabling technology

As consumers become accustomed to more natural sounds, companies need to ensure that the sounds they send are in line with their quality. To be effective, companies must use technologies such as Natural Language Comprehension (NLU), Artificial Intelligence (AI) and Neuroscience-to-Speech (DTS). . Together, these technologies operate on human-like voices, unlike the voices found in today’s intelligent assistants. NLUs, which enable machine reading comprehension, can be used for tasks such as short commands or for more complex tasks such as understanding entire news articles. This communication enables the AI ​​to translate speech and text inputs into their meanings. By integrating these technologies with neurological DTS, companies can create compiled text from human-sounding text and respond appropriately to consumer complaints and questions.

For example, many automakers have embedded voice solutions in their GPS, navigation and telematics systems that can be used for outgoing communication between the vehicle manufacturer and the driver. If an automaker tries to warn a driver about safety risks, they can use a more sympathetic voice. However, if the automaker notifies the dealer of a scheduled service reminder, there is no point in using the same voice for the security alarm. Instead, the automaker can deploy a more upbeat voice when delivering reminders. With the help of NLU, AI and TTS, vehicle manufacturers can make a different voice based on the condition of each consumer.

With the variety of voices they can use, companies can create better, more engaging experiences for consumers that drive loyalty and deepen consumer trust.
balancing the consumer experience

Voice technology is widely used to enhance web access for the visually impaired, and companies in many industries, such as retail, travel and hospitality, are realizing the technology’s benefits at major application events such as GPS devices for vehicles. have been Using it to provide a better, personalized customer experience by interacting with consumers using the different voices of others based on the unique situation of the customer.

Once a voice is created, technology can be modified and measured so that companies can modify scripts, maintain a consistent brand voice, and add new ways to retain consumer attention. That way, companies don’t have to rely on voice actors’ availability or health or want to make adjustments to their custom voices each time they’re recruited, allowing voice actors to be deployed quickly. Once the voice update is complete.

Measurement is important in using different voices with various factors connecting consumers with voice technology. In healthcare, companies may encourage people to exercise, stay hydrated, and take medication, all of which can be stated in an enthusiastic and friendly tone. Companies can remind people to meditate where an upbeat tone doesn’t work. Instead, the company may put in a calming voice in different languages ​​to help users get in the right mood.

As voice technology continues to grow and people continue to demand simplicity and convenience of use, brands today can work to create a digital voice to meet consumer preferences. By using digital voices that capture a brand’s unique personality, companies can better communicate with their most important customers, provide better customer service and

DUBLIN, May 6, 2021 /PRNewswire/ — Soapbox Labs, a leading developer of accurate, safe and secure voice technology for children, today announced the appointment of Dr. Martin Farrows as CEO. For the Irish company it is building a solid share in the education market and expanding into the sports market.

Dr. Patricia Scanlon, who founded the company in 2013 and later served as CEO, will serve as the company’s chief executive and direct future innovation, vision and strategy, with Dr. Farrows serving as its new CEO with immediate effect. Will move on to the role. Dr. Scanlon was named one of the world’s best women in technology by Forbes.

“Soapbox Labs is a firm leader in the education market, but I am pleased to focus on that work as we complete our mission of fundamentally changing the way children interact with technology. in the company’s future technology, vision and strategy,” Dr Scanlan said. Martin is a proven technology leader and operator. His strength, insight and incomparable knowledge of our technology, customers and markets will drive our growth during this important new phase of the company. “

Doctors joined. Faros at Soap Box Laboratory in 2017 as Chief Operating Officer to lead operations, business development and strategic partnerships. His appointment comes as Soapbox Labs makes significant new investments in each of two key offerings: “Soapbox Education” for the existing K-12 market and school-focused market, and “Soapbox Play” for its new consumer-oriented market. . Soapbox Education will continue to develop solutions for voice, language learning, dyslexia screening and speech therapy products, while Soapbox Play will focus on voice-enabled games, toys and entertainment applications for the home and consumer markets.

“With so many new opportunities before us, we are fortunate to have Patricia’s creative and technical prowess that will drive the next phase of our innovation and strategy,” said Dr. Faros said. “I am delighted to be at the forefront of our growth at a turning point. We look forward to rapidly expanding and deepening our leadership position as the voice of choice for the classroom and consumer markets in the years to come.”

Soapbox Labs’ voice operating system powers a variety of gaming and EDEX applications, including a proliferation literacy assessment product and Linguami’s English language learning program. In addition, in 2019, Florida State University’s prestigious Florida Center for Reading Research selected Soapbox Lab for a multi-year partnership to develop and deliver next-generation language and literacy assessments.

In the coming months, Soapbox Labs will announce a number of new deals and client partnerships. The company will also release first-person consumer solutions for applications, toys, games and robots.

As Soapbox Labs expands its market focus, it will also increase its workforce. The company aims to lay off 30 to 80 employees over the next two years, in addition to the 10 new employees it hired last year. It was recently released in the U.S. was issued in. Three I-based employees were raised to leadership positions.

About Soap Box Labs

Soapbox Lab aims to fundamentally change the way kids interact with technology using their voice. Our low-code, independent and proprietary technology provides 95% accuracy in all pronunciations and dialects for children ages 2-12.

The Soapbox Voice Machine was developed using a privacy-oriented approach, and protecting children’s fundamental right to voice data privacy is a cornerstone of our organization’s mission and philosophy.

Soapbox Labs uses the same basic voice engine to address two of our key market propositions:

-Soap Box Education Voice Edec implements to market products and platforms for screening, training, intervention and assessment of literacy and language learning.

Earlier this week, I had the opportunity to interview Clement Simonick, the founder and CEO of Sonyox, who has developed a new AI with its own infrastructure and tools to develop advanced speech and comprehension solutions. With audio to solve complex business problems. “Audio is becoming a widespread medium for fast, high-speed communication,” said Sonyox founder and CEO Clémain Simonic.

According to Grand View Research, Inc., the speech and voice recognition market is projected to exceed $31.82 billion by 2025.

Found in Redmond, California, in April 2020, Sonyox released the first beta release in February 2021, formally launching today, May 13, 2021. Series A will be the next big milestone for the young company. Positioning the market as the most reliable alternative to current voice recognition solutions. There are early signs that they have made significant progress in advancing AI speech recognition solutions.


Sonyox Speech AI uses many anonymous audio and text that are not available to teach itself how to recognize complex speech patterns. As a result, Soniox Speech AI can accurately identify speech in real-world contexts with a better word-error-rate in most human knowledge subjects than with twenty-four percent accuracy. Compared to today’s leading speech systems. Sonic’s mission is to fully understand audio and make it globally accessible and useful.

Kleiman, who has a background in mathematics and computer science technology, immigrated from Slovenia and went to the United States to earn a master’s degree in computer science at the University of Utah. He talked AI on Facebook for more than four years. During this time, he experienced how difficult it was to develop a speech recognition system, thus leaving Facebook’s AI, Clemens’ eighty-year voice recognition team, to create an easy-to-reach product with the most accurate voice recognition system, Brave. Like the entrepreneurs who helped them in their adventure.
more for you
AI 50 2021: America’s Most Exciting Artificial Intelligence
Top 25 Machine Learning Startups to Watch in 2021 Based on CrunchPace
What is the aim of Artificial Intelligence (AI): 2021?

How difficult is it to build an accurate and secure voice recognition system?

Creating accurate and secure voice recognition systems requires significant investment to acquire large amounts of audio-transcribed data, and then have people listen to each audio recording and translate it manually. This process is very time consuming and expensive. More than ten thousand hours of duplicate data must be collected repeatedly to form a rational speech recognition system. Companies like Google and Microsoft have fifty thousand hours of transcript audio. Furthermore, tens of thousands of dollars must be invested in collecting duplicate data. Only then will one be able to practice speech recognition AI from the written data.

Due to the extreme variability of input and output spaces, collecting transcribed data for speech authentication is particularly challenging. Input is interval audio, which can be recorded in any context with a wide variety of background noise, which is easy to analyze and very difficult to translate. For example, there may be multiple speakers with different pronunciations which reduce the simplicity of precise translation.

The place of publication is the sequence of all possible words from any subject of human knowledge. Existing approaches make it (almost) impossible to obtain enough audio-transcript data to cover the complex input and output space.

Sonyox Speech AI technologies have discovered a new approach to training speech recognition models to overcome today’s speech recognition limitations. Using unsupervised AI techniques, Sonyox Speech AI learns from untitled audio and untitled text that are not publicly available on the Internet. It learns to recognize words by examining different interpretations of spoken words in untitled audio and its use in untitled written text.

Now Sonyox Speech AI is unique in recognizing most words in English without any errors, without the need for direct human supervision. The company has a vision to make progress in supporting other languages ​​soon: French, German, Spanish and eventually Asian languages, which are more complex to translate accurately.

What is the market outlook for Voice Recognition?

Voice and audio is the next killer evolutionary application as audio is increasingly pervasive for fast, high-speed communication. Now we see Clubhouse, Twitter Space, Facebook Clubhouse, Spotify, Podcasts and copies of every major technology player: Amazon, Apple, Byte, Facebook, Google, IBM, Salesforce, a voice and emotionally charged AI team, mostly me and Experts in locating with emotion. Other unknown market players include: Deepgram and Red.com, which currently rely heavily on labeled databases against anonymous audio and text, but I suspect they are moving forward with unexpected AI approaches, Given the increasing market dynamics of increasing accuracy and low cost pressure. .

We will continue to see an explosion of solutions that will revolutionize the use of voice and audio and more accurate speech recognition platforms of the future. The exciting thing is that as audio and voice continue to become the medium of communication, the number of users using applications and audio will grow 100x faster in the coming years.

Zoom or more accurately translate all records in Microsoft Teams and get insights.

Some applications that could greatly benefit from the use of accurate speech recognition solutions include automotive automobiles that provide medical advice to distant patients, whether driving a car or turning off or locking doors, and improving call center usage. do.