requests-for-proposal

Real-time Detection of Human Voice in Noisy Environments

Request Number REQ2359817
Due Date December 22, 2017
Request for Proposal Details
RFP Title
Real-time Detection of Human Voice in Noisy Environments
RFP Description

NineSigma, representing a multibillion-dollar manufacturer, seeks a technology for real-time detection of human voice in noisy environments.

Background

Background

The major manufacturer, a client of NineSigma, is engaged in the development of unmanned aerial vehicles (UAVs). Enormous research effort has been made to extract spoken human voice in outdoor environments where UAVs are in use, in order to leverage the voice data for some applications. However, mechanical noise generated from propellers and other noises from surrounding environments make it more challenging.

technology, the real-time detection of the voice is found to be particularly difficult to achieve. In recent years, however, the development of a technology to extract and recognize human voice for various applications, such as mobile devices, has been widely active. Therefore, the client has issued this open request to further accelerate the development of such technology by seeking proposals from potential development partners.

 

Key Success Criteria

Overview of noisy environment and voice to be detected

  • Noisy environment
    • Type of noise: Mechanical noise, sound of rain and wind, etc.
    • Sound pressure: Approximately 100 dB
  • Voice
    • Frequency: Approximately 100 – 1000 Hz
    • Sound pressure: Much lower than the noise; approximately 80 dB

 

Requirements of the technology to be ultimately developed

A technology or system capable to figure out the presence of human voice in the abovementioned noisy environments; the requirements are as follows.

  • Usage environment: Basically outdoor
  • Size: Approximately 10 × 10 × 10 cm at maximum
  • Weight: Should be eventually possible to be a few hundred grams or less
  • Processing speed: In real time
  • Accuracy: Able to determine the presence of human voice in noise; it is not necessary to recognize the content or meaning of spoken words.

 

Proposals need not fulfill these requirements at this moment. Various types of proposals are widely welcome as long as they will possibly meet the requirements within approximately 1 to 2 years of additional development.

 

Possible Approaches

Possible Approaches

Possible approaches might include, but are not limited to:

  • Voice Activity Detection technologies
  • Voice processing technologies using multiple microphones
  • Speech detection technologies using Artificial Intelligence or Machine Learning
  • Frequency filters that can remove noise
  • Selective noise cancelling technologies

Preferred Collaboration types
  • Joint Development
  • Contract Research
  • Technology Licensing
  • Supply Agreement
Items to be submitted

Anticipated Project Phases or Project Plan

Respondents should submit proposals using the attached Response Template.

 

The client will review submitted proposals and possibly ask clarifying questions before selecting the most suitable candidates for collaboration. The client will select the best candidate(s) through evaluations. During the selection process, the client may execute non-disclosure agreements (NDA) with selected respondent(s), seek further information disclosure, and discuss specific development targets or potential opportunities.

The client will execute necessary agreement(s) with the selected respondent(s) and move to the advanced development phase. Specifics of any collaboration will be determined through consultation with the concerned parties.

 

Items to be Included in the Proposal

Responses will use the Proposal Template to include the following items:

  • Overview of the technology
  • Uniqueness of the technology
    • Development stage (Concept level / Performance verified at the lab level / Implemented for practical use)
    • Current performance
  • Applicable environment and conditions
  • Size and weight of the proposed technology or system
  • Processing time
  • Anticipated applications
    • Challenges in fulfilling the ultimate requirements and the future development plans

 

Area of Interest
Request Priority
Comments