Innovative Applications of Speech and Language Technology
Here are some links to innovative uses of speech and language technology. All of these applications are innovative, but no claims are made here about how well they work. If you're interested in learning more about any of these applications, please follow the "more information" links.
1. Using Speech and Language Technologies in Assistive and Rehabilitative Applications
The Application: MossTalk Words
The problem:
Over 1 million Americans have difficulty with language because of an injury to the parts of the of the brain that control language (aphasia).
Aphasia leads to social isolation and inability to work, but insurance only pays for a limited amount of speech therapy.
How speech helps:
A speech application presents a user with aphasia with a picture and the user tries to say the word that corresponds to the picture. Cues are available to help the user remember the word. As soon as the user says the right word, the speech recognizer says “that’s right”. The advantages of using speech technology are:
- Low cost
- 24/7 availability
- Consistency
- Automatic recording of user's interactions with the system
More information
The Application: Animated Speech
The problem:
Millions of preschool and elementary school children have language and speech disabilities but there is a shortage of skilled teachers and professionals to give them the one on one attention that they need.
How speech helps:
The Animated Speech application uses animated agents to produce accurate visible speech. This facilitates face-to-face oral communication and helps in teaching vocabulary to children with language challenges. Instruction is always available to the child, 24 hours a day, 365 days a year. The system has extreme patience --doesn’t become angry, tired, or bored
More information
The Application: Model Talker
The problem:
People who have limited or no ability to speak can use Text to Speech technology to speak their typed utterances, but the number of TTS voices is limited, and there may not be an existing voice that a user likes.
Users who can speak may want to have a computer voice that sounds like their own voice, or a friend or relative's voice that they like.
How speech helps:
Model Talker lets users record their own voice and generate a TTS voice from their own recordings.
More information
The Application: SentenceShaper
The problem:
People with aphasia can have trouble speaking in real time, for example, due to difficulty in word-finding, even though they may actually have considerable knowledge of their language.
How speech helps:
SentenceShaper takes away the real time constraints of creating speech by providing support for recording words and short phrases. It also enables the users to combining their recordings into longer stretches of speech, without requiring the ability to read or use a keyboard.
More information
The Application: Rex the Talking Pill Bottle
The problem:
Some patients can’t read or understand the instructions on their prescription bottles due to illiteracy, low vision, or cognitive deficits.
How speech helps:
A pharmacist can program the bottle with the medication instructions, or the, user can record their own message. Then, to hear the instructions, the user presses a button to hear the instructions.
More information
The Application: Grammar Trainer
The problem:
Children with autism often have trouble learning the grammar of their language the way most children do -- by talking with people during early childhood.
How speech helps:
Some children with autism are oriented toward visual learning, and may benefit from interacting with a structured computer program that assists them in practicing written language.
More information
The Application: iCommunicator
The problem:
Skilled sign language interpreters are hard to find.
How speech helps:
iCommunicator translates speech into American Sign Language and also allows users to type a response that the computer then speaks.
More information
2. Analyzing Speech
Here are some applications that analyze speech for other reasons than recognition.
The Application: Automated Diagnosis of Depression
The problem:
Depression is traditionally diagnosed by self-report, but some depressed individuals are reluctant to admit that their medication is ineffective and so may underreport their depression.
How speech helps:
Depression is correlated with specific vocal characteristics which can be observed in speech recorded over the phone.
More information
The Application: Kish Kish Lie Detector
The problem:
When talking on the phone, you would like to know if the person you're talking to is lying.
How speech helps:
An add-in to Skype measures the stress level in the voice of the person you're talking to.
More information
The Application: Midomi Song Finder
The problem:
You can remember a song but not its name.
How speech helps:
Your singing or humming is analyzed and matched to songs in a database.
More information
|