Wide and Ever Expanding Coverage!
Sensory's speech recognition technologies currently support a wide range of languages covering many countries/regions all over the world.

 Language Coverage Map

We are continuously working to expand our language and country support.
Check back frequently for updated coverage or contact our Sales Department at (971)256-0056 for more information about our language offerings.

Flexible Grammars!
Sensory's Natural Language Interface for the NLP-5x provides the unique ability to understand context-specific user's commands in the natural way the user would like to speak. Order independence allows flexibility in commands and speech prompts can request any missing information (form filling). Revolutionary flexible grammars allow the user to say multiple commands in a single phrase, and even in a flexible order. This results in the most natural use of speech recognition!

Truly Hands-Free Operation
Phrase spotting of up to 30 commands or key words embedded in speech allows the NLP-5x to continuously listen for triggers or commands, even in the presence of high noise. In phrase spotting mode, the word(s) to be recognized may be spoken in the middle of speech. Truly Hands-Free triggers (using phrase spotting) alert the recognizer to listen for commands.

No Training Required!
Speaker-independent (SI) works right out of the box, and requires no end-user training. SI technology is designed for a specific language, and can handle thousands of words in a single set for FluentSoft, 40 words in a single set for the FluentChip/RSC-4x combo, and 75 words in a set for the FluentChip 5/NLP-5x combo. The number of sets is limited only by the amount of memory in your system. With proper design, Sensory's SI technology will yield highly accurate recognition. Sensory's Quick T2SI (text-to-SI) is the first ever GUI tool to allow product designers to create their own speaker-independent set and execute recognition within minutes on chip!

Biometric security through voice
While similar to Speaker-Dependent recognition, Speaker Verification offers the capability of being able to identify whether or not a password is spoken by the original individual who trained that password. The user trains 1-4 passwords (the more passwords, the better the security) that can create voice access to any product. Equal error rates (where the probability of an incorrect acceptance equals that of an incorrect rejection) ranges between 0.01-7% depending on the number of words and whether the passwords are known to the imposter.
On the NLP-5x, up to 10 SV templates can be stored on-chip. The RSC-4x can store 5 SV templates on-chip. With external memory, the number of unique sets for both chips is limited only by programmable memory capacity.

 NanoLock SV Demo

Flexible vocabulary, any language, any accent
Speaker dependent (SD) recognition is desirable where user-specific or language-specific vocabularies are required. Each recognition word is trained just once by the user to create voice "templates", each of which requires up to 200 bytes of memory (which can be on-chip or external). Vocabularies in excess of 100 words are possible, although there are often practical reasons for keeping recognition sets under 50 words. The NLP-5x can store up to 10 SD templates in on-chip SRAM. The RSC-4128 can store up to 7 SD templates in on-chip SRAM. With proper design, Sensory's SD technology can yield highly accurate recognition for any user, regardless of language or accent.

For entering phone numbers and digit strings
This technology is ideally suited for voice dialing applications such as handsets, personal dialers, mobile phones, and hands-free kits. It can also be used anywhere that a string of digits are used for recognition.

Text-based Speech Playback
Text-to-speech (TTS) is supported for systems requiring text-based speech playback, and requires less than 1MByte of external memory. TTS works well for names or text/phrase reading and is supported in multiple languages. Sensory’s Hybrid Speech can compress speech playback data even lower.

Hi-Fidelity Stereo MP3 Decoder with all standard bitrates and a 5-band equalizer.

Sensory's music synthesis technology can produce up to 24 stereo voices simultaneously at a sample rate of 32K samples-per-second on the NLP-5x. The RSC-4x family supports up to 8 mono voices simultaneously at 8KHz sample rate. The music can be played through the on-chip stereo DAC or mono PWM (Pulse Width Modulator). Speech or sound effects encoded with SX, PCM, or ADPCM can be mixed in with the music. The music synthesis technology can "play" MIDI files that are stored in on-chip or off-chip memory, such as serial flash. MIDI files are a memory-efficient way to store music. The music synthesizer requires a database of instrument audio samples which is typically stored in external parallel flash memory. Sensory currently offers a database of instrument samples for a wide variety of common instruments from the General MIDI melodic instrument set, plus the complete General MIDI percussion set.

Perfect for Voice Prompts and/or Speech Output
High quality speech and sound effects can be played back by Sensory’s IC’s and software products. Sensory's compression technology utilizes proprietary time and frequency-domain approaches that can compress speech and sound effects to as little as 1000 bits per second. Speech output creates the opportunity for natural dialog with a product and can reduce reliance on an instruction manual.

Store messages and play back--voice messaging capability
Compressed digital sound reproduction
Sensory's RSC-4x and NLP-5x processors can record audio to off-chip RAM or Flash at data rates of under 30k bits per second for custom greetings, phones and answering machines, voice pitch changers, and hand-held recording devices. On-chip compression levels can be varied depending on the quantity and quality of playback desired. Automatic silence removal can also be done to reduce memory requirements. The NLP-5x offers 8k and 16k bit samples per second while the RSC-4x family offers 8k samples per second. The NLP-5x signal processing provides superior voice quality.

LCD control logic and drive - up to 104 icons or pixels. SPI for large array driver interfaces.

Motor control logic - up to 3 bi-directional motors.

Silent SonicNet communicates data via encoded sound at 14KHz or 18KHz in short bursts on the NLP-5x. These high frequencies make the short bursts essentially inaudible in practical application. Silent SonicNet can run conincident with SX or T2SI, allowing data transmission during VR dialogues. Products with integrated speech that already include an NLP-5x, microphone and speaker can implement this at no additional cost, and can interact with each other, potentially doubling demand.

Communicates data at 8KHz via encoded sound in short bursts on the RSC-4x. SonicNet can run coincident with SX to partially mask the sonic tones. Products with integrated speech that already include an RSC-4x, microphone and speaker can implement this at no additional cost, and can interact with each other, potentially doubling demand.
 SonicNet Demo

USB1.1, SPI, UART-Lite, I2S and infrared (IR) interfaces combine with voice user interface capabilities, enabling man-machine interface solutions with an unprecedented combination of power and cost-effectiveness.

Allows a product to match robotic mouth movements to speech heard in real time, much like a ventriloquist dummy.
 Real-Time LipSync Demo

Chip figures out the recurring beat to know how to act moving forward-great for dancing and motion oriented products.
 Beat Prediction Demo

Allows for a product to match robotic mouth movements to pre-recorded speech.
 LipSync Demo

Picking up the amplitude of different sounds in the room as they occur and reacting to them with a movement or display function.
 

A human pitched voice can be analyzed by the RSC processor to figure out the pitches being sung.

Combining talkback and pitch detection allows a robotic creature or avatar to imitate a person singing.
See the Windows Media demo

This demo features a dog barking and matching pitches from a human voice.
See the Windows Media demo

Adding a second microphone allows the RSC-4x processor to locate the direction of a human voice.
 SoundSource Demo

The RSC can produce speech in response to your talking or inquiries that appears to be conversational speech from a non-human creature.

Set radio stations using natural phrases on the NLP-5x.

Set digital time clocks using natural phrases.
 Natural TimeSet Demo

Wake from low power mode capability
One of the challenges for hands-free battery operated products, was that if they were always on, always listening the batteries would drain rapidly. Sensory has created a low power technology on the RSC-4x that can listen for audio (whistle or claps) and wake up from this low power mode and begin listening for speech recognition commands. This technology can extend the life of battery operated products from weeks to years.

Sensory and 3rd party developers provide support for presence detection, touch and position sensors, gesture and motion analysis, etc. USB1.1, SPI, UART-Lite, I2S and infrared (IR) interfaces combine with voice user interface capabilities, enabling man-machine interface solutions with an unprecedented combination of power and cost-effectiveness.

Speech Recognition and TTS for Headsets, Music Players, Hands-Free Kits & More
Sensory’s BlueGenie Voice Interface software suite runs on CSR's BC-5 MM Kalimba DSP, and enables manufacturers of Bluetooth products to integrate full voice control and synthetic speech output without the need for visual displays or complex user interfacing. It frees designers to pack functionality onto small form factor Bluetooth devices and answers consumer demand for a truly hands-free experience. TTS allows Caller ID announcement and SMS message playback with speech.
 BlueGenie Voice User Interface Demo