2

I'm looking for prior art for US patent application 2011/0054890 filed 2009/08/25.

That would be a mobile phone (or portable device) that is capable of detecting the direction and type of sounds with the ability to display the results. I've found some art for gun shot detection, but nothing as general as this where the sounds, according to the specification, can include: laughing, crying, traffic, etc.

The representative independent claim and the salient refinements are:

1. An apparatus comprising: at least one processor and at least one memory including computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus at least to:

  • identify at least one audio type based on a signal representative of sound;
  • determine the direction of any of said identified audio types; and
  • provide feed-back to a user of the identified audio types and the direction of said audio types.

2. An apparatus according to claim 1, further comprising at least one audio input, wherein said signal representative of sound is received via said at least one audio input.

3. An apparatus according to claim 2, wherein said at least one audio input port is at least one microphone.

5. An apparatus according to claim 1, further comprising a motion sensor, and said processor uses an output from said motion sensor to determine the direction of the at least one audio type.

6. An apparatus according to claim 1, further comprising a digital compass, and said processor uses an output from said digital compass to determine the direction of the at least one audio type.

Improve this question
4
  • Please, always provide the patent or application number, and a link to the patent page on Google or a patent office such as USPTO.
    – Gilles 'SO- stop being evil'
    Commented Oct 22, 2012 at 20:34
  • This is impossible! You can't determine the direction of sound with only one microphone, as on a cell phone! It takes 3 points to triangulate a position in a plane, and 4 for 3D space.
    – Chloe
    Commented Oct 23, 2012 at 5:05
  • @Chloe A quick glance through the patent indicates that the user must manually rotate the mobile device 360 degrees for the "direction detection" algorithm to identify the direction of the sound. As such, it seems it will detect literally only the direction of the sound, not the position of its source.
    – kinkfisher
    Commented Oct 25, 2012 at 3:58
  • @Chloe "impossible" might be too strong a word. You can make a pretty good estimation of the 3D location of a sound with only 2 microphones (your ears). Particularly if the sound source or your head is in motion. A single microphone with an asymmetrical baffle would be able to make some broad but reasonably accurate assumptions about the location of a moving sound source or a static source of known audio qualities.
    – Fracture
    Commented Jul 26, 2013 at 21:08

4 Answers 4

Reset to default
2

This may not be close enough, but it is pretty much the same thing in a military application.

https://www.qinetiq-na.com/products/survivability/ears/

SWATS accurately detects gunfire and determines its source location in less than a second. Users are alerted to a threat by audio and/or visual reporting

This pretty much covers claim 1.

Improve this answer
2

Learning Sound Location from a Single Microphone by Ashutosh Saxena and Andrew Y. Ng, presented at ICRA 2009 in May 2009, anticipates claims 1–3, albeit not by the one-year period required in the US to be indubitably prior art.

1. An apparatus comprising: at least one processor and at least one memory including computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus at least to:

It is not explicit in Saxena that a computer is used to perform the method. However, Saxena discloses the use of a sensor containing a microphone (Fig. 3, Fig. 4, §IV.A. Saxena discloses the use of a learned hidden Markov model (learned HMM) applied to the data obtained via the microphone (§IV.B). It would be obvious to one skilled the art to perform the learned HMM on a computer comprising a processor and a memory.

  • identify at least one audio type based on a signal representative of sound;

Saxena, title; Saxena, §IV.B (“Our test data comprised a variety of different sound sources: (…)”). Saxena is able to distinguish between a chosen audio type and background noise (§IV.B, “Our data was collected in a normal office environment, which had some background noise (…)”).

  • determine the direction of any of said identified audio types; and

Saxena, §IV.B, p.4–5 (in particular “note a recurring pattern in the transfer function that changes as a function of the direction”); Saxena, Fig. 6.

  • provide feed-back to a user of the identified audio types and the direction of said audio types.

Saxena, Fig. 4; Saxena, Fig. 5. Saxena discloses providing feedback in a form that is amenable to computer scientists studying sound localization; it would be obvious to one skilled in the art to provide feedback that is more suitable for a user by displaying the highest-confidence option.

2. An apparatus according to claim 1, further comprising at least one audio input, wherein said signal representative of sound is received via said at least one audio input.
3. An apparatus according to claim 2, wherein said at least one audio input port is at least one microphone.

Saxena, Fig. 3 discloses an audio input port that is a microphone.

Improve this answer
1

There is nothing new or novel about most of this patent, acoustic direction finding has been around since the 1800's and was used to detect approaching aircraft pre radar http://www.douglas-self.com/MUSEUM/COMMS/ear/ear.htm . This device is doing exactly the same thing but using modern electronic devices devices. The bit about identifying the type of sound by comparing it to stored profiles is not a particularly novel idea either the technique is used in electronic warfare to identify target radars - the detected signal is analysed by a computer and compared to a stored library. Their technique for analysing the sound maybe novel.

Improve this answer
-1

I cannot comment on any specific application or anything related there to, however, I can question in general whether audio "types", identified or not, have "direction". I can say with some certainty that they do not in this universe. You might want to think about that one for awhile, and also consider 35 U.S.C 112 1st paragraph, there is a part of it that deals with impossible things.

You can also cite, against any such apparatus, in general, which supposedly comprises a 1 or a 0 therein any of the off the shelf computers and note that any of the 1's or 0's inside are just such a configuration as is called for by the claim.

Bottom line, don't take language attempting to distinguish a computer in terms of 1's and 0's too seriously.

Likewise, in claims generally about 1's and 0's if you have the government granted power to do so, make them show you such a configuration, that is, the physical configuration, in the drawings. Just so you understand exactly which collections of 1's and 0's you are discussing when you're talking about "configurations". And then consider how many different arrangements of 1's and 0's they have sufficient written description for in their spec. If they have none, they probably lack written description under 35 U.S.C 112 1st paragraph.

Improve this answer
6
  • 1
    Even though a patent's claims are the most important part, they are still to be read in conjunction with the rest of the patent. You only need to read the abstract to figure out that what they mean by "direction" is the bearing of whatever is emitting the sound...
    – vt.
    Commented Oct 23, 2012 at 16:38
  • "what they mean by "direction"" What "they" mean by direction, in a hypothetical case which doesn't exist (remember I cannot comment on any given specific case), is irrelevant. What "they" claimed is relevant, in a hypothetical case that doesn't exist. The broadest reasonable interpretation must include what they literally claimed as well as what you one might think they "meant to say". Ex Parte Miyazaki helps resolve such issues generally. A simple 112 2nd paragraph will get them to put what they "meant" or "meant to say" into the claim. At which time you can reevaluate if they have WD.
    – 6answersugenerally
    Commented Oct 23, 2012 at 17:19
  • And after you reevaluate if they had Written Description, you can lol. At least this is what I do, and it brings me much joy when I deal with cases which think they can tots distinguish their apparatus in terms of invisible (and imaginary 99.9999% of the time) 1's and 0's that they didn't even bother to draw. The first step is getting them away from having original claims still present in the application. It is much harder making a 112 1st paragraph case out against original claims.
    – 6answersugenerally
    Commented Oct 23, 2012 at 17:22
  • Anecdotal evidence, but having read the description and being skilled in the part, I think I could definitely implement what they are describing. Regarding what the "direction" of an "audio type" means and whether the claim distinctly identifies the invention, I think it's irrelevant here given that the question asked was specifically about prior art. It would be trivial for them to go back and use a different term like "isolated sound" but any prior art would still stand.
    – vt.
    Commented Oct 23, 2012 at 17:59
  • Also, could you please write like a human? It took me longer to understand your comments than the patent itself.
    – vt.
    Commented Oct 23, 2012 at 18:01

You must log in to answer this question.