Available Technologies

No technologies match these criteria.
Schedule UC TechAlerts to receive an email when technologies are published that match this search. Click on the Save Search link above

Find technologies available for licensing from UC Berkeley.

Monolithically Integrated Implantable Flexible Antenna for Electrocorticography and Related Biotelemetry Devices

A sub-skin-depth (nanoscale metallization) thin film antenna is shown that is monolithically integrated with an array of neural recording electrodes on a flexible polymer substrate. The structure is intended for long-term biometric data and power transfer such as electrocorticographic neural recording in a wireless brain-machine interface system. The system includes a microfabricated thin-film electrode array and a loop antenna patterned in the same microfabrication process, on the same or on separate conductor layers designed to be bonded to an ultra-low power ASIC.

RF-Powered Micromechanical Clock Generator

Realizing the potential of massive sensor networks requires overcoming cost and power challenges. When sleep/wake strategies can adequately limit a network node's sensor and wireless power consumption, then the power limitation comes down to the real-time clock (RTC) that synchronizes sleep/wake cycles. With typical RTC battery consumption on the order of 1µW, a low-cost printed battery with perhaps 1J of energy would last about 11 days. However, if a clock could bleed only 10nW from this battery, then it would last 3 years. To attain such a clock, researchers at UC Berkeley developed a mechanical circuit that harnesses squegging to convert received RF energy (at -58dBm) into a local clock while consuming less than 17.5nW of local battery power. The Berkeley design dispenses with the conventional closed-loop positive feedback approach to realize an RCT (along with its associated power consumption) and removes the need for a sustaining amplifier altogether. 

MyShake: Earth Quake Early Warning System Based on Smartphones

Earthquakes are unpredictable disasters. Earthquake early warning (EEW) systems have the potential to mitigate this unpredictability by providing seconds to minutes of warning. This warning could enable people to move to safe zones, and machinery (such as mass transit trains) to be slowed or shutdown. The several EEW systems operating around the world use conventional seismic and geodetic network infrastructure – that only exist in a few nations. However, the proliferation of smartphones – which contain accelerometers that could potentially detect earthquakes – offers an opportunity to create EEW systems without the need to build expensive infrastructure. To take advantage of this smartphone opportunity, researchers at the University of California, Berkeley have developed a technology to allow earthquake alerts to be issued based on detecting earthquakes underway using the sensors in smartphones. Called MyShake, this EEW system has been shown to record magnitude 5 earthquakes at distances of 10 km or less. MyShake incorporates an on-phone detection capability to distinguish earthquakes from every-day shakes. The UC Berkeley technology also collects earthquake data at a central site where a network detection algorithm confirms that an earthquake is underway as well as estimates the location and magnitude in real-time. This information can then be used to issue an alert of forthcoming ground shaking. Additionally, the seismic waveforms recorded by MyShake could be used to deliver rapid microseism maps, study impacts on buildings, and possibly image shallow earth structure and earthquake rupture kinematics.

Lockout Tagout Software

Energy Isolation Lock out Tag out (“LOTO”) is a series of CalOSHA and FedOSHA code compliance requirements and is the primary means by which equipment must be rendered “safe” prior to allowing personnel to work on the equipment.  LOTO codes require equipment-specific written procedures identifying all types of energy sources needed to operate the equipment as well as the energy-isolation methods and locations of utility disconnects, stored energy, etc. In addition, every LOTO procedure must be annually verified to confirm the written procedure is still accurate to the equipment.   Whereas current LOTO procedures are typically hand-written or using other time-consuming processes, UC Berkeley authors have created software allowing users to retrieve LOTO procedures in real-time guiding the end-user through a logical thought process to allow them to identify all energy sources and safety processes, and equipment needed.  

Zero-Quiescent Power Transceiver

Trillions of sensors are envisioned to achieve the potential benefits of the internet of things.  Realizing this potential requires wireless sensors with low power requirements such that there might never be a need to replace a sensor’s power supply (e.g. battery) over the lifetime of that device.  The battery lifetime of wireless communications devices is often governed by power consumption used for transmitting, and therefore transmit power amplifiers used in these devises are important to their commercial success.  The efficiencies of these power amplifiers are set by the capabilities of the semiconductor transistor devices that drive them.  To achieve improved efficiencies, researchers at UC Berkeley have developed a novel method and structure for realizing a zero-quiescent power trigger sensor and transceiver based on a micromechanical resonant switch.  This sensor/transceiver is unique in its use of a resonant switch (“resoswitch”) to receive an input, amplify it, and finally deliver power to a load.  This novel technology also greatly improves short-range communication applications, like Bluetooth.  For example, with this technology, interference between Bluetooth devices would be eliminated.  Also, Miracast would work, despite the presence of interfering Bluetooth signals.

Novel Interactive System for Collective Insight Generation & Visualization

The volume of interactions in social media and crowd-sourcing tools continues to significantly grow.  As the amount of data being shared increases, showcasing relevant information has become a significant challenge.  Many social networking sites use linear lists for online discussions and crowd-sourcing feedback. Unfortunately, these systems do not scale well.  One major problem with linear lists is the amount of data presented to an end-user can become overwhelming. As an example, if a particular news story generates thousands of responses, then this data is impractical to navigate using a linear list, biases users to whatever data is presented at the top, and impedes consideration of the diversity of responses.  To address this situation, researchers at the University of California, Berkeley have developed a novel method to interactively visualize data for an online environment. This system can be applied to responses that are in textual, numeric and multimedia formats. By using canonical correlation analysis and other techniques, researchers have been able to highlight the most relevant information for end-users and in turn, facilitate browsing, and rating of responses, as well as displaying informative patterns.

Low Capacitance/High Speed Bipolar Phototransistor

The performance of optoelectronic links is very strongly related to the sensitivity of the detector on the receiver end. Conventional receivers include a photodiode whose signal is sent to amplifiers until it is strong enough to be used in microelectronic circuits. The energy cost of amplification is very high and could be significantly reduced if the capacitance of the photodiode and first stage of amplification were smaller. In order to be useful for this application, a phototransistor must have several features: - Low capacitance - High speed - Large photon absorption volume Unfortunately, for conventional bipolar phototransistors, these requirements are contradictory. Indeed the photon absorption length in typical semiconductors is on the order of microns, while the speed requirement only allows transit regions for amplified carriers of a few tens of nanometers at best. This is over a 100x size mismatch. Increasing any other dimension (that is not the transit direction) results in prohibitively high capacitances. This invention offers a solution to these issues consisting of a new kind of semiconductor phototransistor device, which integrates a large PIN-photodiode with a bipolar junction transistor (or Heterojunction Bipolar transistor). 

Enhanced Patterning Of Integrated Circuits

Information and communication technologies rely on integrated circuits (ICs) or “chips.” Increased integration has improved system performance and energy efficiency, and lowered the manufacturing cost per component. Moore’s Law predicts that the number of transistors on an IC will double every two years, yet industry experts predict that we are reaching economic limits of traditional circuit patterning processes. Photolithographic patterning is best suited to print linear features that are evenly spaced. The smaller or more complex the shape, the more likely the printed pattern will be blurred and unusable. Although multiple-patterning techniques can be used to increase feature density on ICs, they bring a high additional cost to the process. This means that the most advanced ICs available today have a high density of features, but are restricted to having simple patterns and are increasingly expensive to produce. Without innovations in production techniques, Moore’s Law will reach its end in the near future.  To address this issue, researchers at UC Berkeley have developed a one-step method to increase feature density on chips. This method is capable of achieving arbitrarily small feature size, and self-aligns to pre-existing features on the surface formed by other techniques. 

A New Coding Technique For Interference Mitigation

Background For high data rates and massive connectivity, the next generation cellular networks are expected to deploy many small base stations. While such dense deployment provides the benefit of bringing radio closer to end users, it also increases the amount of interference from neighboring cells. Consequently, smart management of interference is becoming a key enabling technology for high-spectral-efficiency, low-power, broad-coverage wireless communication. Over the past decades, several techniques at different protocol layers have been proposed to mitigate adverse effects of interference in wireless networks. One important conceptual technique at the physical layer is simultaneous decoding whereby each receiver decodes for the desired signal as well as part or whole of interference. When interference is strong this simultaneous decoding technique achieves the optimal performance for the two user Gaussian interference channel using good point-to-point codes. Moreover, it achieves the optimal maximum likelihood decoding performance in general, when the encoders are restricted to point-to-point random code ensembles. The celebrated Han-Kobayashi coding scheme, which achieves the best known performance for general two-user interference channels, also uses simultaneous decoding as a crucial component. As a main drawback, however, each receiver in simultaneous decoding has to employ some form of multiuser sequence detection, which usually requires high computational complexity to implement.   Technology Description Engineers from the University of California have developed a low-complexity coding foundation for communication channels with multiple pairs of senders and receivers, in which the signals from the senders interfere with each other and thus the signal observed at each receiver is a mix of the desired signal as well as one or more interfering signals and some noise. This technology will mitigate the adverse effect of interference caused by other communicating parties. More specifically, this technology decomposes a data stream into multiple substreams. These substreams are communicated over multiple units (“blocks”) of the span of time/frequency/space dimensions. Each sender encodes each of its substreams into a codeword that spans over multiple blocks and transmits multiple codewords simultaneously by superimposing them in a staggering manner. The characteristics of the codewords (coded modulation) and the mechanism of superimposing them (superposition) can be optimized with respect to the communication channel parameters as well as other transmission constraints. Each receiver recovers the codewords from its desired sender as well as some codewords from interfering senders by decoding its received signal over a sliding window of multiple blocks. For each window, multiple codewords (both desired and interfering) can be recovered one by one (successive cancellation decoding), which allows for each decoding step to be low-complexity. The selection of the codewords to be recovered as well as their decoding order can be optimized.

A Network-Connected, Low-Power Early Warning Device For Natural And Man-Made Disasters

Earthquake early warning (EEW) networks are prevalent in several earthquake prone nations. For example, the Japanese EEW network has provided seconds to minutes of warning across the country - saving lives and properties. These EEW networks make use of the ability of sensors near a rupture point to transmit information about the rupture faster than the propagation of the earthquake wave. This is analogous to how the observed delay between a lightening flash and the corresponding thunder clap increases with distance from the lightening location. Likewise, the time delay between the EEW warning and an earthquake shaking can increase with distance from the epicenter. In 2013 the State of California mandated the development of an EEW system. However, the State hasn't funded the full deployment of the systems, so it is only available as a beta system in selected areas for selected entities.To leverage this emerging EEW, researchers at UC Berkeley (who have access to this EEW beta system) have developed an EEW alarm that has similar characteristics to ubiquitous, consumer fire/smoke alarms. The distinguishing attributes of the Berkeley earthquake alarm include: a low cost of manufacturing; easy installation (i.e. it doesn't require a professional to install); the form-factor of a home fire alarm; wireless connection to the EEW network; low power (3V, 5V or power-over-internet); battery back-up; always-on operation; audible and visual alerts. The device could also potentially connect to other warning networks such as for tsunamis, tornadoes, chemical spills, radioactive fallout, civil unrest, air-raids, etc. For detailed information, go to: http://5nf5.blogspot.com/2014/09/early-warning-device-of-earthquakes-and-other-maladies-for-everyone.html

Hybrid Porous Nanowires for Electrochemical Energy Storage

Supercapacitors are attractive energy storage devices due to their high-power capabilities and robust cycle lifetimes.   “Super” capacitors are named in part because the electrodes are composed of materials with high specific surface area and the distance between the electrode and electrochemical double layer is very small compared to standard capacitors.  A variety of porous silicon nanowires have been developed for use as supercapacitors electrodes by maximizing the specific surface area of active materials.  Although the use of Si is attractive due to its wide-spread adoption by microelectronics industry and due to its abundance, Si nanowires are highly reactive and dissolve rapidly when exposed to mild saline solutions.  Previously, silicon carbide thin films were used to protect the porous silicon nanowires, but the coatings were 10’s of nm thick and while they successfully mitigated Si degradation during electrochemical cycling in aqueous electrolytes, they also resulted in pore blockage and a large decrease in energy storage potential.   Researchers at UC Berkeley have developed methods and materials to improve porous silicon nanowires by overcoming the above limitations.  The resulting nanowires have an ultrathin carbon coating preserving the pore structure while mitigating Si degradation.  The resulting supercapacitor electrodes have the highest capacitance (and hence energy storage) per projected area to date.   

Dynamic Searchable Encryption with Minimal Data Leakage

Dynamic Searchable Symmetric Encryption (DSSE) methods are used to perform searches on encrypted data without decryption. This allows for the secure storage of data on remote or third-party storage while maintaining the ability for the user to access data as needed. However, current DSSE schemes have drawbacks that make practical implementations of DSSE difficult. Some methods produce data leakage, where information about what is being stored can be learned by the storage server via access patterns, hashes of encrypted search keywords, keywords shared among different documents, access to deleted documents, total number of documents stored, and other information. Other implementations provide stronger security against data leakage but are impractical because of search time, additional storage space required, or a chance of false positive occurring from a search.Researchers at the University of Maryland’s Maryland Cybersecurity Center (MC2) and the University of California Berkeley have developed a dynamic searchable encryption method that minimizes the amount of data leaked when performing a search query. The method allows for a sublinear search time in the worst case while guaranteeing that the only information that can be leaked to the server is the access and search patterns and deleted documents that match the searched keyword. The server requires a roughly linear amount of storage, and the client requires only a small amount of storage at any time (in the hundreds of megabytes). Adding or deleting a keyword to search requires a minimal amount of bandwidth (kilobytes). This method has been implemented in a cloud storage environment (Amazon EC2) and been shown to be practical with a variety of database sizes, network latencies, and results per query.

Wireless High-Density Micro-Electrocorticographic Device

A minimally invasive, wireless ECoG microsystem is provided for chronic and stable neural recording. Wireless powering and readout are combined with a dual rectification power management circuitry to simultaneously power to and transmit a continuous stream of data from an implant with a micro ECoG array and an external reader. Area and power reduction techniques in the baseband and wireless subsystem result in over 10x IC area reduction with a simultaneous 3x improvement in power efficiency, enabling a minimally invasive platform for 64-channel recording. The low power consumption of the IC, together with the antenna integration strategy, enables remote powering at 3x below established safety limits, while the small size and flexibility of the implant minimizes the foreign body response.

Silicon Photonic Switch

Optical (photonic) switches are used in datacenter networks to switch signals carried by optical fibers from one circuit to another. They deliver reconfigurable bandwidth, which improves network performance and reduces power consumption. Currently, optical switches are limited either by their slow switching time (ms) or small switch size as well as high cost, which restricts their use in data-intensive applications. Investigators at UC Berkeley have developed a photonic switch that can be monolithically integrated on a silicon chip. It has a fast switch time (microsecond), large port count (~ 100 or larger), and low optical loss. Furthermore, the switch can be mass-produced at a low cost, allowing for easy scale up. These advances will be particularly beneficial in high traffic datacenter networks that require fast switching times. A prototype of the switch has been experimentally demonstrated at Berkeley Marvell Nanolab.

Micromechanical Frequency Divider

Frequency dividers have become essential components of phase-lock loops and frequency synthesizers that are used in a variety of applications from instrumentation to wireless handsets. In a typical frequency synthesizer application, frequency dividers often limit the achievable phase noise performance and contribute a large or even majority portion of the total power consumption. Common digital dividers offer good noise performance, but at the cost of power far in excess of that permissible for mobile applications and with poor scaling as frequency is increased. To alleviate this, injection-locked oscillator dividers have emerged as low power options at high frequencies, but they have performance limitations due to the active transistors used to sustain oscillation.To overcome these limitations, researchers at UC Berkeley have developed a new design for frequency dividers. While performing a frequency divide-by-two function, a version of this on-chip MEMS-based frequency divider reduced phase noise by 6 dB at close-to-carrier frequencies and 23 dB far-from-carrier. Unlike conventional frequency dividers, this Berkeley design dispenses with active devices and their associated noise, and operates with close to zero power consumption, limited in principle only by the power required to overcome MEMS resonator loss, estimated at 100 nW. With an output voltage swing of 450 mVpp generated from only 445 mVPP of input swing on a version of this MEMS divider, cascaded chains of fully passive dividers are possible, as needed for use in real-world phase-lock loops and frequency dividers. 

Dynamic Proof of Retrievability from Cloud Storage

Data storage outsourcing has become one of the most popular applications of cloud computing, offering benefits such as economies of scale, flexible accessibility, efficiency, and allowing companies to focus on their primary business activities. Due to the increase in percentage of services conducted online and number of mobile internet connections, demand for data storage continues to grow. Customers in this industry are primarily concerned with authenticated storage and data retrievability. Although many efficient proof of retrievability technologies have been developed for static data, only two dynamic technologies exist. However, both are too expensive to implement in practice due to the fact that they require a high level of bandwidth. To address this problem, researchers have developed a dynamic proof of reliability scheme that requires 300 times less bandwidth than currently available technologies. This innovative technology makes dynamic proof retrievability of data practical and efficient, and thus attractive for the industry implementation. This technology gives clients of cloud storage providers assurance that their data has not been modified and that no data loss has occurred.

Piezoelectric Filter with Tunable Gain

There is a long-standing problem of how to switch piezoelectric filters when used in switchable filter banks -- such as needed in RF channel-selection. To address this problem, researchers at UC Berkeley have developed a method and structure for a piezoelectric resonator with tunable transfer function -- i.e. tunable gain. This Berkeley resonator's gain is tunable to many values -- including values that are low enough to consider the device to be "off" relative to the background signal. Accordingly, this approach enables on/off switching of piezoelectric resonators; and it thereby obviates the need for separate low loss switches, which otherwise would be needed in series with piezoelectric resonators to switch them on and off -- adding insertion loss and raising system gain. In addition, this ability to adjust filter gain makes it possible for the resonator to control low power gain in a receiver front-end.

MEMS-Based Charge Pump

The reduction of power supply voltage with each new generation of CMOS technology continues to complicate the design of charge pumps needed for high voltage applications that integrate into systems alongside transistor chips -- such as the increasing number of MEMS-based gyroscopes, timing oscillators, and gas sensors. Moreover, the aggressive scaling in CMOS resulting in lower dielectric and junction breakdown voltages has compelled the use of customized CMOS processes -- including increased gate oxide thickness and/or added deep-n-wells. Clearly, advances in transistor technology are moving in the opposite direction of the needs of high voltage MEMS applications. To address this trend, researchers at UC Berkeley have developed a MEMS-based charge pump. This design avoids the turn-on voltage and breakdown limitation of CMOS. With much higher breakdown voltages than transistor counterparts, the demonstrated MEMS charge pump implementation should eventually allow voltages higher than 50V desired for capacitive-gap transduced resonators that currently dominate the commercial MEMS-based timing market.

MEMS Ultrasonic Fingerprint ID System

Two-dimensional optical fingerprint analysis has been used for a variety of personal identification applications over the years.  However, automated optical fingerprint scanning techniques have a number of limitations that block their use in broader applications. For example, automated optical fingerprint scanning techniques sense only the epidermal layer of a fingerprint. As a result, they are prone to errors created by finger contamination.  The marketplace has reflected the limitations of optical fingerprint identification, as many optical fingerprint scanners have been removed from most later models due to these limitations. They lacked the necessary robustness to perform predictably in such everyday environments.  Ultrasonic fingerprint scanners have been developed in an effort to minimize the limitations of currently available optical fingerprint scanning, and avoid some of the resulting errors.  However, currently available ultrasonic fingerprint scanners devices are limited in their applications because of large size, the requirement of a physically moving scanning device, and cost.   UC researchers have developed a micro-machined ultrasonic transducer fingerprint identification system (MUT fingerprint ID system) to address these issues.  MUT fingerprint ID system has advantages of a small size, robust solid-state construction, easy fabrication, easy integration with electronics, and fast electronic scanning. These features represent a game-changing advancement over currently available bulky, failure prone mechanical scanners.  The system also has orders of magnitude lower cost per unit than current systems.  Conventional fingerprint sensors used in consumer electronics applications are capacitive sensors and are extremely prone to errors due to wet, dry or oily fingers. Optical sensors are sensitive to dirt on fingers. Unlike both capacitive and optical sensors, which measure the fingerprint on the epidermis (skin surface), the ultrasonic sensor at the core of the MUT fingerprint ID system can detect the fingerprint on both the epidermis and dermis (subcutaneous) layers.  

Systems and Methods for Electrocorticography Signal Acquisition

Systems and methods for biosignal acquisition, and in particular, electrocorticography signal acquisition, are disclosed for small area, low noise recording and digitization of brain signals from electrode arrays.

Permanent Magnet Flux Loop Linear Generator/Actuator

There is increasing commercial interest in small-scale, electricity generator applications that harvest energy from mechanical vibrations or linear motion.    To address this interests, researchers at UC Berkeley have developed a magnetic circuit architecture that has higher flux densities -- on the order of one Telsa -- across large functional air gaps. This circuit generates large induced voltages that can be easily rectified and stored to power wireless devices such as condition monitor sensors.    This innovative circuit can be used to efficiently transduce any linear kinetic energy but is particularly attractive for small-scale applications because the magnetic circuit generates large induced voltages for overall device length scales on the order of millimeters and centimeters. The source of the kinetic energy that is transduced can come from coupling to mechanical motion, mechanical vibration, current carrying conductors, fluid flows or pressure differences.

First Practical ORAM for Concealing Access Patterns to Data on the Cloud

Many organizations and individuals encrypt data that they store in the cloud to achieve confidentiality and privacy. However, when this data is accessed interactively (such as through a cloud storage service like Dropbox or Skydrive), this encryption isn't enough to ensure privacy. By observing the locations of the accessed data, attackers can often easily recover information about the encrypted data without ever needing to decrypt it. To address this problem, researchers at UC Berkeley have developed Oblivious RAM (ORAM) software for securely concealing a client's access patterns to data residing in a cloud environment. This enables files to not only be encrypted, but it also prevents attackers (and even the cloud service provide itself) from determining which files (or portions of files) the client is accessing. Furthermore, the Berkeley ORAM software contains techniques for achieving practical performance under realistic scenarios as well as reducing network latency and memory requirements. This development includes a framework that is extensible and readily combined with other algorithms. 

Concave Nanomagnets With Widely Tunable Anisotropy Properties

Nanomagnets form the basis of the nanomagnetic logic field, in which data computation is achieved using magnetic field instead of electrical current. Circuits using nanomagnets would replace transistors in computers. There are many advantages of this technology, including small size and low power requirements. In order to achieve functional nanomagnet circuits, however, the nanomagnets themselves must be tunable.Researchers at UC Berkeley have developed nanomagnets with concave geometries that can be used to achieve widely tunable anisotropy control. The tuning is in both direction and strength, making this invention useful in a wide variety of magnetic device applications.

Decoding Heard Speech And Imagined Speech From Human Brain Signals

Thousands of severely disabled patients are unable to communicate due to paralysis, locked-in syndrome, Lou Gehrig’s disease, or other neurological disease. Restoring communication in these patients have proven a major challenge. Prosthetic devices that are operated by electrical signals measured by sensors implanted in the brain are being developed in an effort to address this problem.  Investigators at University of California at Berkeley have responded to this challenge by developing an algorithm to decode speech, including arbitrary words and sentences, using brain recordings from the human cortex.  A computational model is trained and determines how recorded electrical signals at specific brain sites represent different speech features, for example acoustic frequencies.  The trained model then takes as input novel brain recordings and outputs a set of predicted speech features.  Once these steps are accomplished, speech sounds are either directly synthesized or words are identified from the predicted speech features using statistical techniques.  The brain signal decoding algorithm can decode speech solely from brain signals and may permit communication via thought alone.   

Modulation-Domain Speech Filtering For Noise Reduction

There are problems with discrimination speech from background noise in applications such as cell phones, hearing aids, telecommunications, automatic speech recognition, which limit the use of these important technologies. In response to this challenge, investigators at University of California at Berkeley have developed a speech filtering algorithm that improves intelligibility of speech in noisy backgrounds.  This speech filtering algorithm is applied as a first stage of processing, before transmission, analysis, or listening.  The speech filtering algorithm can be located in its entirety on a dedicated DSP or FPGA.   These systems allow the complexity needed for the computations of the gain stage, but still function fast enough for real-time denoising. The investigators have implemented the speech filtering algorithm using a computer to process signals offline (post-hoc). The inspiration for our model comes from our research on the statistics of animal vocalizations (including human speech) and on the responses of auditory neurons to these sounds. They have shown that neurons in avian primary and secondary auditory areas are tuned to represent the statistical structure of vocalizations. These areas are analogous to cortical auditory areas in humans.  The preferred embodiment uses many relatively narrow bands in the analysis filter-as low as 50 Hz bandwidth and spacing. Another embodiment uses fewer, wider bands in the analysis filter, perhaps several thousand Hz in width.The most significant advantage of X is its ability to exploit important, basic statistical information about all human speech in a simple fashion. Some of these properties are included implicitly in advanced ASR systems. Separating these simple, physically-driven features from more complicated verbal and syntactic rules brings this power to bear in simpler, lower-power systems like cell phones. Additional complexity is also the primary drawback of this system: it requires more computational power, and because it uses a longer history, it could introduce delays that are unacceptable for low- like hearing aids and cochlear implants.

  • Go to Page: