Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

TPT17

No description
by

Millecamp Martijn

on 18 March 2016

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of TPT17

Pitch?
piptrack
Results
TPT17 Multimedia Indexing and Retrieval
Pitch Detection
Question Detection

Preprocessing
-zero padding
peak stays
smoother
better performance
Sinusoidal Peak Estimation
QIFFT
Quadratic interpolation
Conclusion
Our task was to discover a question in the sequence of the audio track based on the prosody.
We were able to understand the feature of pitch and it's importance.
In our case this is the only feature we looked at
We successfully generated pitch representation.
We failed at finishing up the project on time.

Questions
non-questions
Feature Extraction
Main feature: pitch

Window size: 30 ms, with overlap of 10 ms

Vector size corresponding to 15 windows

So, the training example base corresponds to (30-10)ms * 15 = 300 ms of data
Labeling
Questions and non-questions


* A supervised approach


Gaussian Mixture Model
The GMM presents itself as a possible solution considering the absence of a precise parameter that could be useful to spot a question.
Time domain
zero crossing
autocorrelation
Frequency domain
Cepstral analysis
Find Peak
librosa.piptrack
Prosody
properties beyond those described by the individual segments, which include
speaking rate
timing
pausing
articulatory quality
voice quality
pitch

Pitch?
Pitch is an auditory sensation in which a listener assigns musical tones to relative positions on a musical scale based primarily on their perception of the frequency of vibration.

Pitch is closely related to frequency

pitch is perception of a sound, can be percieved also subjectively
frequency is an objective, scientific attribute that can be measured
However, the number of components would be critical, once the question intonation is not an easy-to-track property.
Full transcript