Common Voice
Why is this project worthy of an award?
Common Voice is a voice data bank and speech recognition service that aims to collect all of the world’s languages and dialects to make it easier for anyone to get online using the power of their voice. It’s making voice data publicly available for anyone to train machine learning speech models, and ensuring it represents true linguistic and speech diversity. When people speak to Alexa, Siri or Google Voice, the interactions are logged, but the ever-expanding set of voice data that results is under proprietary control — unavailable to startups, researchers, and noncompetitive companies. Since these entities can only access limited data sets — which are costly, insufficient in scale, and unavailable in many languages — innovation and economic diversity suffer. The open and publicly available Common Voice database of audio files allows anyone to use it to train new voice-enabled applications. People donate their voices through an iPhone app or at https://voice.mozilla.org/ . We combine this with data we’re compiling and curating from other publicly available sources. We package the data so it’s appropriate for training machine learning-enabled speech applications. And anyone can freely download it under a Creative Commons CC0 license, without attribution or restrictions In the first five months we’ve collected and released the initial tranche of donated voices: nearly 400,000 recordings, or 500 hours of speech. In addition, we’ve had contributions from more than 20,000 individuals, including people from Bangladesh, France, Germany, India, Indonesia and dozens more countries, reflecting a diversity of voices globally. Speech is becoming a preferred way to interact with personal electronics, saving time and allowing the development of technologies without a screen. Unlike most existing speech recognition engines which prioritize socially advantaged, North American and European, and male voices based on the data on which they are trained, Common Voice starts with the broadest possible database of voices to limit inherent discrimination. This will result in a greater diversity of voices — opinions, points of view, ideas, and insights — joining global online culture. Common Voice is an initiative from Mozilla, the not-for-profit dedicated to ensuring that the internet is a global public resource, open and accessible to all. Our position as a not-for-profit allows us to make decisions for the good of individuals and the ecosystem of the internet versus corporate shareholders and investors. As new ways of accessing the internet emerge — from virtual reality to speech recognition and beyond — Mozilla works to expand access and opportunity as a baseline for technology development that benefits the greatest number of people. It’s likely that Common Voice open-source data will become part of the voice banks for all voice-assisted technologies in the future. By working together in the open across the globe, we will influence the world’s largest technology companies to expand a welcome to the internet’s knowledge, opportunity, and connection to everyone.
What else would you like to share about your design? Why is it unique and innovative?
The way we are building Common Voice is through a global community that includes volunteers, academic institutions and researchers, governments, developers, startups and established companies. Each is contributing their piece to create a high-value public resource that all can benefit from and that none could build on their own. For example, given the size, quality and availability of this data set, it can be used to create the benchmark for judging the performance of machine learning speech applications. Current benchmarks are flawed, and so we’re working with a number of companies and academics to jointly establish the characteristics that a quality benchmark should have. And we will release it together. In another example, we are working with companies and people who are passionate about their languages to expand Common Voice to those parts of the world. The Common Voice app is itself open sourced, which allows these communities to localize and extend the functionality in a way that is locally relevant. And finally, we are working with universities and governments to incorporate their data to make the project more successful, which involves adjusting licenses and cleaning up the data or its structure. For example, we worked with researchers at the University of Illinois to make the Flickr30k data set available through CC0 licensing, giving users of the Common Voice app more than 130,000 new sentences to read. Common Voice complements Mozilla's work in the field of speech recognition, which runs under the project name "Deep Speech" (https:/github.com/mozilla/DeepSpeech). Together with a community of like-minded developers, companies and researchers, we have applied sophisticated machine learning techniques and a variety of innovations to build an open-source speech-to-text engine that has an accuracy approaching what humans can perceive when listening to the same recordings. Together with the growing Common Voice data set we believe this technology can and will enable a wave of innovative products and services, and that it should be available to everyone. Common Voice was among the best open-source rookies of 2018, awarded by InfoWorld magazine.
Who worked on the project?
Michael Henretty, Open Innovation, Mozilla Kelly Davis, Machine Learning, Mozilla
View the project video: https://www.youtube.com/watch?v=VJrxFFW5G-c
CONNECT: A Mega Skyport
Company Corgan
Introduction Date May 8, 2018
Project Website https://www.corgan.com/story/connect-corgans-mega-skyport/
Why is this project worthy of an award?
Today’s lifestyle moves at a rapid pace, but our daily commute prohibits maximum efficiency. Corgan’s CONNECT concept for the Uber Mega Skyport embraces urban air mobility, ushering in a new paradigm in transportation. Designed to meet the anticipated aerial ridesharing demand for uberAIR service, CONNECT supports electric vertical takeoff and landing vehicles (eVTOL) through a flexible and modular station that is scalable to meet massive or episodic passenger throughput in a way that encourages community acceptance. The idea of urban aerial transportation is closer than most people think and CONNECT offers a modular skyport solution that can be constructed today and grow as demand increases. Within the dense urban environment, the networks of highways and rail lines that have left major divides in the urban fabric also contain the aerial real estate for the “skylanes” needed make this concept a reality. While initial single module skyports can easily be placed on parking garages or rooftops, larger paired modules are designed to span these urban divides and reconnect our communities. Co-locating the paired modules with existing infrastructure not only maximizes land use and optimizes multi-modal connectivity, it also masks aerial mobility activity within established noise corridors, which is key to achieving community acceptance. The design of CONNECT embraces the importance of the community in developing a Skyport that promotes an aerial ridesharing lifestyle. The Connection Plaza is the iconic pedestrian entry to the Skyport, connecting all modes of transit. As the heart of the Skyport, the Station not only provides physical connectivity across the highway but also serves as a destination hub itself with shops, restaurants, parks, and performance centers to offer attractive amenities to neighbors and riders alike, while reflective of the community’s culture. At the center of each module is a monumental circulation core that moves passengers from the Station to the Flight Deck. The Flight Deck tops the Skyport and is where all aerial activity occurs. Openings around the cores provide a visual connection to the Station below and allow the cores to serve as multi-story guideposts, providing a vital connection between the community and flight. The CONNECT Uber Mega Skyport represents a new lifestyle of mobility. Ready for immediate implementation and flexible to accommodate increasing demand, CONNECT opens up the possibilities of technology to shape a new culture of urban aerial mobility.
What else would you like to share about your design? Why is it unique and innovative?
Anticipating the demand and democratization of uberAIR, CONNECT overcomes many of the technical challenges associated with peak usage and activates the Skyport as a vibrant mixed-use space 24-hours a day while embodying the following design principals to envision a Skyport built for this generation and designed for the future: Modular and Scalable: The existing network of highways provides a robust framework for uberAIR to integrate into existing infrastructure. Initially, uberAIR will serve communities through a single CONNECT module via ground or rooftop stations. As demand increases, CONNECT’s single module can be paired, stacked, and scaled up to a Mega Skyport to handle a higher volume of operations Intuitive User Experience: It is important that passenger movement through the Skyport is as intuitive as Uber’s passenger interface today. CONNECT is crafted so that whether you are at the pedestrian level or flying through the sky, the Skyport serves as a guidepost and facilitates direct and intuitive passenger movement while still providing the opportunity for the user to extend their experience and explore all that the Skyport has to offer. Convenience and Connectivity to Community: As Uber introduces air travel into their offerings, CONNECT will reinforce and enhance the customized, personalized offering provided today. By integrating into existing infrastructure networks, CONNECT’s centralized connectivity elevates operations to higher levels of service and convenience for everyone—supporting mass adoption and linking daily commuters and visitors alike to established transit networks, Uber car services, and public transport. Noise Control: Although uberAIR vehicles are relatively quiet, it is important to manage the impact of noise throughout cities and neighborhoods. CONNECT minimizes added intrusion and conceals noise by locating Skyports within compatible use zones and existing noise corridors above highways. Because of the distance between flight operations and people, noise at ground level is reduced to that of a standard electric vehicle and becomes increasingly negligible as CONNECT continues to increase in height above the surrounding Station and Connection Plaza environment. Sustainability: Strategically locating the CONNECT system above otherwise unusable airspace and near existing infrastructure allows for easy integration to mass transit networks while preserving ground space for higher value transit-oriented development adjacent to the Skyport. To reduce reliance on the public power grid and support vehicle charging, CONNECT takes advantage of traditional sources of self-generation and harvests downwash from uberAIR operations and converts it into a high efficiency charging source. Green space within the Station adds park-like amenities that also reduce the heat island effect and collect and control rainwater runoff. While the primary purpose of CONNECT is to serve as a transportation hub for uberAIR service, Corgan seized the opportunity to connect neighborhoods and create individualized experiences. The vision for CONNECT is to address the greater needs of the community and take advantage of the unutilized space found above highways in every major city. By bridging the divides highways have left in the urban fabric, the design has the potential to reconnect communities and provide a commercial and social opportunity to serve the neighborhood.
Who worked on the project?
John Trupiano, Principal, Corgan Samantha Flores, Associate, Corgan Scott Gorenc, Senior Associate, Corgan Keith O’Connor, Vice President, Corgan
View the project video: https://vimeo.com/267432898
Cora’s Reimagined Pads for Periods and Light Incontinence
Why is this project worthy of an award?
Cora’s reimagined pads are the first major innovation in the stagnant menstrual pads market in decades. Launched in 2016, Cora provides women with modern methods to manage their periods while simultaneously working towards period equality around the world. For every month’s supply of Cora purchased, the company provides a month’s supply of pads to a girl in need. The seed to create the reimagined pad was planted two years ago after Cora saw an incredibly positive response to the company’s organic tampons and subscription UX. As the team explored existing pads on the market, even new brands were offering antiquated, diaper-like pads. After rigorous customer research, it took the Cora team a full year to bring the company’s innovative pad to market. During that process, Cora also learned from its customers the need to create a product for light bladder leakage given that 40% of women over the age of 18 report experiencing regular light bladder leaks. While almost half of women experience light bladder leakage, only 25% use an incontinence product. Many women turn to period pads and liners to manage the issue, however those pads aren’t built for incontinence. With that in mind, Cora set out to not only create a better period pad, but also the first hybrid pad for both periods and incontinence. The pad’s modern shape and patent-pending channels maximize absorbency. It also uses the most cutting edge new materials, making the pad 50% thinner than other ultra-thin pads, and eliminating more than 20 toxic chemicals found in typical pads. The pads themselves are only .9mm thick and absorb 11x their weight. In regards to the material science behind the pads, what makes them so unique is the innovative mix of the highest performing materials possible without using concerning chemicals found in regular pads - this is because of the way that the materials within the pad interact with one another. The topsheet, while cashmere soft, is engineered specifically to take liquid away from one’s body and immediately distribute it down into the pad. Unlike cotton pads or many other materials, Cora’s top sheet will not hold liquid, creating a dry surface against one’s body. Additionally the pad has channels that distribute liquid across the entire pad (in comparison to traditional pads which only absorb in the middle third of the pad). The channels only work if they are truly connected through the depth of the pad and can’t simply be “embossed” into the pad. This meant that Cora also needed to redesign the manufacturing process to properly incorporate these channels given that the line produces over 250 pads per minute. Lower down in the pad is a layer of absorbency technology that converts the liquid into a gel. Because of the channels, this liquid to gel transformation happens throughout the entire length of the pad. Additionally, Cora’s period pads include side edges and wings that are hydrophobic and help prevent liquid from traveling from the center of the pad off the edges thereby preventing leaks.
What else would you like to share about your design? Why is it unique and innovative?
Cora’s reimagined pads for periods and light bladder leakage are unique and innovative for reasons beyond the products themselves. In the same way Cora has helped to dismantle the stigma of periods, the company created The Cora Hybrid Pad to address an ultra-taboo issue that affects one in three women under 50: light bladder leakage. Previously, light bladder leakage has been portrayed as something experienced only by older adults, even though it's often experienced by younger women and can be brought on by any number of the natural changes a woman's body will go through such as pregnancy and giving birth, increased physical activity, or body mass changes. Cora wanted to address the issue of bladder leakage in a way that let women on the younger end of the spectrum know that we don't believe they're experiencing bladder leaks because their bodies are broken or shameful, but because they're incredibly powerful, and that this experience can be easily managed without indignity or embarrassment . By showing younger women they are not alone in their experience, and creating a modern, design-led solution that women can use during their periods as well as for light bladder leaks, we are helping to remove the shame women have felt about their bodies. Additionally, the company’s giving model for pads and liners is the same as it is for all other products: with every month's supply of products delivered to a Cora subscriber, the company provide a month's supply of period pads and reproductive health education to a girl in India through our partner, Aakar Innovations, and with every product purchased at Target, Cora gives 10% of profits to provide pads to girls in Kenya through its partner, Zana Africa.
Who worked on the project?
Molly Hayward, Co-Founder, Cora Morgen Newman, Co-Founder, Cora
View the project video: https://drive.google.com/drive/u/0/folders/10pybDD3bh3P_7fLz1JzDOMP_5CBvXaVz
Could Silicon Be Quantum Computing’s Wonder Material?
Company Intel
Introduction Date February 14, 2018
Project Website https://newsroom.intel.com/news/intel-sees-promise-silicon-spin-qubits-quantum-computing/
Why is this project worthy of an award?
Quantum computing is an exciting new computing paradigm with unique problems to be solved and new physics to be discovered. It has the potential to tackle problems conventional computers can’t handle. For example, quantum computers may simulate nature to advance research in chemistry, materials science and molecular modeling – like helping to create a new catalyst to sequester carbon dioxide, or create a room temperature superconductor or discover new drugs. While there’s been significant progress, quantum computing research is still nascent. The industry is at mile one in a marathon, and to realize this new computing paradigm, many problems must be solved and many architectural decisions must be made. For example, it’s not yet clear what form quantum processors (or “qubits”) will take. That’s why Intel is placing two major research bets and investing in them equally. One possible form is superconducting qubits. Intel is making fast progress in developing this type of test chip, which others in the industry and academia are also pursuing. Additionally, Intel is researching an alternative structure, which draws on the company’s world-class expertise manufacturing silicon transistors. This alternative architecture is called “spin qubits,” which operate in silicon and could help overcome some of the scientific hurdles to take quantum computing from research to reality. Spin qubits resemble the semiconductor electronics and transistors as we know them today. They deliver their quantum power by leveraging the spin of a single electron on a silicon device and controlling the movement with tiny, electrical and microwave pulses. Electrons can spin in different directions. When the electron spins up, the data signifies the binary value 1. When the electron spins down, the data signifies the binary value 0. But, similar to how superconducting qubits operate, these electrons can also exist in a “superposition,” which means they have the probability of a spin that’s up and down at the same time and, in doing so, they can theoretically process tremendous sets of data in parallel, much faster than a classical computer. For qubits to operate as designed, they need to operate at extremely cold temperatures in order to prevent noise or unintended observation, which can cause data loss. These very cold temperatures (colder than deep space) create challenges for the material design of the chips themselves and the control electronics required to make them work. In contrast to their superconducting counterparts, spin qubits offer several advantages in addressing the challenges of bringing quantum computing to commercial reality – particularly in relation to managing the fragile nature of the qubits themselves and enabling technologists to identify new ways to scale the system to the larger qubit counts required for a commercially viable solution. Recently, Intel’s industrial partner, QuTech, demonstrated success creating an industry-first two-qubit spin-based quantum computer. This can be programmed to perform two simple quantum algorithms. This development paves the way to the larger spin-based processors that Intel is fabricating today and will be capable of more complex applications and, ultimately, a new potential path to quantum computing.
What else would you like to share about your design? Why is it unique and innovative?
QuTech’s two-qubit spin-based processor represents the first step on the path to the development of larger processors capable of more complex applications. And, spin-based quantum computers provide the engineers at Intel some advantages in scaling the quantum system over their superconducting counterparts. They’re small and strong: Spin qubits are much smaller in physical size and their coherence time is expected to be longer – an advantage as researchers aim to scale the system to the millions of qubits that will be required for a commercial system. They can function at higher temperatures: In additional experiments, Intel and QuTech have shown that silicon spin qubits can operate at higher temperatures than superconducting qubits (1 kelvin as opposed to 20 millikelvin). While still very cold, this could drastically reduce the complexity of the system required to operate the chips by allowing the integration of control electronics much closer to the processor. Intel and academic research partner QuTech are exploring higher temperature operation of spin qubits with interesting results up to 1K (or 50x warmer) than superconducting qubits. Intel manufacturing know-how: The design of the spin qubit processors highly resembles the traditional silicon transistor technologies. While there are key scientific and engineering challenges remaining to scale this technology, Intel has the equipment and infrastructure from decades of fabricating transistors at scale. And, Intel has already invented a spin qubit fabrication flow on its 300 mm process technology using isotopically pure wafers sourced specifically for the production of spin-qubit test chips. Fabricated in the same facility as Intel’s advanced transistor technologies, Intel is now testing the initial wafers. Within a couple of months, Intel expects to be producing many wafers per week, each with thousands of small qubit arrays.
Who worked on the project?
Intel • Michael Mayberry, Intel Corporation • James Clarke, (Components Research, Intel Corporation) • Hubert C. George, Components Research, Intel Corporation • Kanwal Singh, (Components Research, Intel Corporation) • Ravi Pillarisetty (Components Research, Intel Corporation) • Nicole Thomas (Components Research, Intel Corporation) • Jeannette Roberts (Components Research, Intel Corporation) • Hubert George (Components Research, Intel Corporation) QuTech • Lieven Vandersypen (QuTech and Kavli Institute of Nanoscience, TU Delft) • Luca Petit (QuTech and Kavli Institute of Nanoscience, TU Delft) • Jelmer Boter, QuTech and Kavli Institute of Nanoscience, TU-Delft • Nima Kalhor (QuTech and Kavli Institute of Nanoscience, TU Delft) • Nodar Samkharadze (QuTech and Kavli Institute of Nanoscience, TUDelft) • Amir Sammak (QuTech and Netherlands Organization for Applied Research) • Diego Sabbagh (QuTech and Kavli Institute of Nanoscience, TU Delft) • Giordano Scappucci (QuTech and Kavli Institute of Nanoscience, TU Delft) • Menno Veldhorst (QuTech and Kavli Institute of Nanoscience, TU Delft) • Raymond Schouten (QuTech and Kavli Institute of Nanoscience, TU Delft) • Ruoyi Li (QuTech and Kavli Institute of Nanoscience, TU Delft)
View the project video: https://youtu.be/_ImX9mUwHa0
Coverage
Company PSF Projects Architecture
Introduction Date November 12, 2017
Project Website http://www.psfprojects.com
Why is this project worthy of an award?
Coverage was the winning entry to the Van Alen Institute’s Emergency Contacts competition which took place in November 2017. The Van Alen Institute sought “proposals for an archetypal public mental health treatment center that will provide an imaginative street-level presence for a range of mental health services integrated within local communities to address the gap in access to mental healthcare, and help destigmatize mental illness.” Van Alen Institute juror Karen Kubey, a well-known urbanist and healthcare specialist, noted on behalf of the jury that, “Coverage is the winning entry because we felt we were looking at something that both thought strategically about de-stigmatizing mental health and how these things are integrated in the city and also gave a really strong argument for the role of design.” We are grateful to the Van Alen Institute for inspiring us to engage design to address a very real social problem. For this project our architecture firm, PSF Projects, collaborated with an experienced mental health clinician, Annie Reed, to ensure our concept was grounded in the realities and needs of actual mental health practice. We have since elaborated on our initial competition proposal to develop the design more completely. Many online and digital resources exist for mental health support, but these are often invisible to a person in need. Our modules and their distribution in the city would also be digitally accessed and make these resources evident in two ways. First, potential clients can find a QR code linking to a local online mental health support website mounted to the module. Second, they can locate the nearest module by finding them on a live tracking application. Our next step is to identify a partner organization with whom we can collaborate to bring this concept to fruition in the form of a physical prototype. The exposure Fast Company might give our proposal could be a significant help to us in securing the necessary partners and funding to bring this concept for an improved mental healthcare system to the populations in need of better care.
What else would you like to share about your design? Why is it unique and innovative?
What if mental health resources were affordable, readily available and easily replicated? Designed with NYC in mind, our proposal can be deployed in other cities and university settings. With the deinstitutionalization of mental health services the number of state psychiatric hospital beds plummeted from over 550,000 nationally in the 1950s to less than 40,000 today. Today, tragically, in a mental health crisis people are more likely to encounter police than get medical help. In addition mental health trends in the United States warrant a revised broad-based approach, as suicide rates and opioid related deaths have been increasing since the year 2000, and studies show that more than half of Americans report feelings of loneliness. We harness the power of design to create a network of immediately identifiable treatment locations which are wood-clad modules that can be deployed in several ways. Our strategies to best reach the populations in need, both Medicaid and private pay, include providing a broad range of services in de-stigmatized, self-supporting environments, with the ultimate goal of improving care and reducing ER visits and psychiatric admissions. We know that access to a medical doctor or psychiatric nurse practitioner to provide timely blood testing, EKGs, and prescriptions is key to efficiently addressing the treatment needs of Medicaid patients. To this end we propose a dispersed network of radially distributed contact points. The hubs of this system, perhaps one to three per borough will be commercial testing labs where an MD or NP will consult with new patients, order tests performed on site, and write prescriptions. The typical 4-week lag between initial contact and medication receipt for new patients will be drastically reduced, and a sustainable revenue stream for the facility will be provided, through rent from the lab and professional offices. Surrounding these hubs will be neighborhood treatment centers, located in street-level storefronts. These centers will house modular therapy offices, readily deployable and reconfigurable as needed. The modules will function as spatial organizers, defining areas for reception, group therapy, harm reduction and 12-step programs, and mindfulness and meditation classes. The therapy offices will also be available for rent to private practices, to broaden the services available and provide supplemental revenue. Dispersed at a finer grain yet through the catchment area of each hub will be stand-alone mobile modular kiosks serving both outreach and intake functions, located in parks, public housing sites, and other public spaces. These modules will be aggregated as crisis centers during terror attacks, natural disasters, or other events that might leave neighborhoods distraught or even destroyed. The tectonic and graphic identity of the treatment access points will be consistent and recognizable across the three scales of intervention. For example, the modular therapy rooms at the core of the treatments centers would also serve as phlebotomy rooms in the lab hubs and - in a weatherized version - as the outdoor kiosks. The material language and form we envision for the modules utilizes natural materials and maximizes daylight to provide a calming environment for healing.
Who worked on the project?
Barrett Feldman, Principal Jonah Pregerson, Principal Mark Sofield, Principal Juan Tejedor, Designer Evan Ortiz, Designer Annie Reed, Mental Health Clinician, LCSW
View the project video: https://vimeo.com/269055104
