Questions tagged [speech-processing]

Speech processing is the study of speech signals and the processing methods of these signals.

Filter by
Sorted by
Tagged with
9
votes
1answer
1k views

How does noise reduction for speech recognition differ from noise reduction that is supposed to make speech more “intelligible” for humans?

this is a question that has interested me for some time now, mainly because I'm working on noise reduction for an existing speech recognition system myself. Most papers on noise reduction techniques ...
8
votes
3answers
7k views

How do I go about detecting whistles, pops and other sounds in live audio input?

I've read many questions on SO, and frankly, each one of them is not describing any particular way to go about it. Some say "do FFT" and some say "zero crossing" etc. But I've only gone as far as ...
6
votes
1answer
2k views

How do I construct input to neural network from audio signals?

Input: Microphone recordings of digits from 0 to 9 from different speakers. Output: The digit from 0 to 9. I am doing this for fun. So first I will train my neural network using some samples and ...
5
votes
2answers
10k views

Calculation of Reverberation Time (RT60) from the Impulse Response

I have some confusions regarding reverberation time (RT60). I need to calculate reverberation time from a given power envelope. This is what I get. As you might see the line is curved close to zero (...
5
votes
1answer
21k views

Why should one use windowing functions for FFT?

So I just revised my pitch calculation algorithm using a harmonic product spectrum algorithm. I was just curious about why this explanation of Harmonic Product Spectrum states that you need to ...
5
votes
1answer
4k views

Voice Audio Detection algorithm

I have to detect speech intervals in large pre-recorded files in my project. I think, there won't be so much noise in the background (audio will be recordered in the room or even in the studio), but ...
4
votes
1answer
754 views

Speech recognition: words vs. phonemes

One could train a speech recognizer with words or phonemes. I guess that they both have their pros and cons: with words there would be a lower error rate as phonemes can be put together building a ...
4
votes
2answers
313 views

Matching LPC frequency response to raw sample data frequency response

Background: The intention is feature/formant extraction from speech input. I have used an FFT on raw samples in the past and have done simple peak picking from a weighted average of frequencies over ...
3
votes
3answers
529 views

Signal processing for audio and speech

I have started learning about signals and I am interested in sound signals. There are some questions that I need to resolve. For example in music, difference of notes like 'sol' and 'la' are about ...
3
votes
2answers
5k views

Why is each window/frame overlapping?

I am currently reading up on some speech recognition theory. As it speech signals usually are non stationary, but if a speech is seen in a very small time frame can it be deemed stationary, as would ...
3
votes
2answers
2k views

Where can I get a database of speech signals?

For a classification sort of experiment using Matlab, I am in need of a database of male and female voices. So where can I get links of such open databases.
3
votes
2answers
111 views

Is There a Sparse Representation for Noise?

Is there sparse representation for stationary noise and nonstationary noise? How can I learn dictionary for each noise class? (my mean of noise is noises with which speech signals are often ...
3
votes
2answers
2k views

what does musical noise sound like and is there a sample clip for it?

Musical residual noise is created when a noisy signal passes through a noise reduction algorithm. For instance, please see the paper: Esch, T; Vary, P: "Efficient Musical Noise Suppression for Speech ...
3
votes
2answers
843 views

Algorithm to detect if a spoken phrase matches a previously spoken phrase

I would like to create a very simple system that allows a user to train (once or multiple times) a spoken phrase (from 1 word to a whole sentence.) Then the (same) user would speak the phrase back, ...
3
votes
1answer
144 views

How to prevent Octave Jumping in ACF of Speech Signal?

I'm working with speech signals and my aim is to estimate the fundamental frequency $\ F_0$ of this signal often called as "pitch". The main idea is taking small blocks of the speech signal such ...
3
votes
0answers
912 views

Pre-emphasizing in speech recognition

Usually, in speech recognition, the techniques that are used are based on the linear prediction model (Fant, 1960). Parallel to this, the human speech production mechanism causes energy to drop ...
3
votes
0answers
9k views

MFCC calculation

I've got a project and need to calculate MFCCs. I tried to read some tutorials and then make a MATLAB function but I seem to have wrong answers. I'd like to feed MFCCs to one of the classification ...
2
votes
2answers
146 views

What do cloud-based speech-to-text services upload to the cloud?

Services like Siri and Google Now perform speech recognition (speech-to-text) on mobile phones by doing part of the computation on a server in the cloud. What do they upload to the cloud? Do they ...
2
votes
2answers
1k views

Difference Between MFCC of speech and speaker recognition

I am doing my project on speaker recognition. One of my friends is doing his project on speech recognition. Both of us need to calculate the MFCC for feature extraction. Is the MFCC going to be the ...
2
votes
1answer
1k views

Question about overlap-add

@Jim Clay: No if the filter varies slowly, then the clicks go away. So I guess the filter needs to vary slowly to avoid audible clicking sounds. I created a test in matlab where I'm comparing time-...
2
votes
1answer
584 views

Can't compute real cepstrum of real signal

I'm trying to compute the real cepstral coefficients of recorded telephone audio in Matlab using the rceps function. On some audio frames (480 samples per frame -- 60ms of audio at 8kHz), I get a ...
2
votes
1answer
61 views

Linear Predictive Coding - Meaning of spectrally Flat

I am a student with very little background on signal processing and am studying the Linear Predictive Coding algorithm. I came across the term spectrally flat in ...
2
votes
1answer
237 views

Parseval's Theorem for discrete series

I need to use Parseval's Theorem to find \begin{equation} S\:=\:\sum _{n=-\infty }^{\infty }\:\left[\left(\frac{\sin\left(\frac{\pi }{4}n\right)}{2\pi n}\right)\left(\frac{\sin\left(\frac{\pi \:}{6}n\...
2
votes
1answer
119 views

Cepstrum calculus disambiguation

Learning the cepstrum analysis for speech recognition, I have met two different definitions of cepstrum (for discrete signals): $F^{-1}(ln|F(x[n])|)$ . That is, the cepstrum is the inverse Fourier ...
2
votes
1answer
163 views

Using 8 bit samples for speech comparison

I am making a speech comparison app in javascript. I found this dsp.SE post: Speech comparison algorithm for rating on similarities, where I read it is best to use mfcc and dtw algorithms. I ...
2
votes
1answer
54 views

How to correctly classify voice from other sound events?

I have a huge labeled dataset of several thousand sound events, including human voice, dish washing, things falling to the ground, among others. I need to report when a human voice event takes place....
2
votes
1answer
219 views

Autoregressive modeling (linear prediction) of electrical transmission lines?

I've read that the reflection coefficients in speech processing (as computed by the Levinson-Durbin algorithm for solving the Yule-Walker equations) "represent the fraction of energy reflected back" ...
2
votes
3answers
1k views

Speech corpora for VAD

I'm doing a student research project connected with VAD. And I want to find speech corpora with added noise. I found many references to the NIST Speaker Recognition Corpora. But unfortunately, all ...
2
votes
1answer
1k views

Detecting phones in a word audio

I am working on speech recognition for Hindi language. My approach is to map speech to text based on phones. For this, first split the speech into the words and then map each word to text by detecting ...
2
votes
1answer
838 views

Dividing Audio signals into Barks Bands

I want to apply the following steps in matlab 1) Read the audio data 2) Apply window function on that data 3) Apply fft on the result of window function 4) Make power spectrum of fft 5) divide ...
2
votes
1answer
2k views

Log energy calculation

I have some code here that I don't quite understand. It calculates the energy for a given frame the following way: En=log(1+energy(32767*frame)) The energy ...
2
votes
1answer
109 views

How to detect time of arousal with excitation signal in speech signal?

I have several seconds of speech signal and I have to detect moments in time of arousal with excitation signal, which is creating by larynx. I'm looking for useful reference since several days. I know ...
2
votes
1answer
67 views

How to reduce synthesis artifacts produced by phase vocoder?

I implemented the phase vocoder algorithm in Python to time-stretch speech signals by following this paper and referring to this MATLAB tutorial. I can distinguish words in the original signal from ...
2
votes
0answers
221 views

What are F0 counters in Speech signals? What is represented by F0?

I can't find any dedicated definition for Frequency F0 counters. Can anyone tell me what does F0 exactly mean, I am confused about that.
2
votes
1answer
332 views

MFCC Classification

I'm working on gender estimation from speech signal and I completed MFCC feature extraction. So now I'm trying to estimate gender from these features. But I have frames for an audio file and I ...
2
votes
0answers
115 views

Speaker diarization tool for acoustic analysis

I'm researching certain acoustic prosodic features during live clinical interactions. Because of the setting, the recordings always have cross channel bleed, and in some cases everything is recorded ...
2
votes
1answer
162 views

Speech signal formant spectrogram. How can we determine F1?

The formant spectrogram in this page points F1 and F2 as below; In bottom-right F1 movement, I thought F1 could be below blue one. But how can we determine F1 is upper one not below one? Is it ...
2
votes
0answers
94 views

Cepstrum of finite impulse train

In Acero's Spoken language processing (p. 309) it says that the cepstrum of a finite length impulse train $$x[n] = \sum^{M-1}_{k=0} \delta[n-kN]$$ is given by $$\hat{x}[n] = \sum^{ \infty}_{r=1} \frac{...
2
votes
0answers
222 views

How to interpret the sign of a cepstral coefficient?

I feel like I've got a decent intuition for what the magnitude of a cepstral coefficient represents (i.e. how much of the signal's energy is contained in harmonics of a given wavelength), but I don't ...
2
votes
0answers
177 views

Speech features that can be inverted?

I am trying to figure out speech features that can be inverted to recover the audio back with no loss or as less loss as possible. I am currently using the power spectrum which is completely ...
1
vote
2answers
283 views

How to edit audios so that they have the same length

I have a dataset of audio files, which differ on their lengths by a couple of milliseconds. Some are little shorter than 1s and some are little longer than 1s. They all have the same sampling rate, ...
1
vote
2answers
565 views

Why discrete cosine transform may not maintain locality

In paper Convolutional neural network for speech recognition, they say I didn't understand the highlighted sentence.
1
vote
1answer
2k views

In framing of audio samples', what is need of frame shift while giving frame size??

In framing of audio samples' in audio feature extraction, what is need of frame shift while giving frame size?? i.e. frame size = 20ms, frame shift = 10ms. Rather then shifting/overlapping why ...
1
vote
2answers
1k views

When should we using moving average in algorithm design?

I'm new to signal processing. And I just read G.720.1. I find there is a lot of moving average used. And I can also recall there is a lot of moving average been used in other audio processing ...
1
vote
1answer
6k views

MFCC Deltas, Delta-Deltas

I have just a quick question about how to calculate Deltas from MFCC. I have found the formula for calculating deltas here: $$d_t=\frac{\sum_{n=1}^{N}n\left(c_{t+n}-c_{t-n}\right)}{2\sum_{n=1}^{N}n^2}...
1
vote
1answer
155 views

Logarithmic Spacing

What is the meaning of sampling logarithmically? How do you space the samples logarithmically? Or to rephrase the second question, How to sample an audio signal logarithmically? It would be ...
1
vote
1answer
28 views

Is there any way that we can perform speech recognition without using Fourier transforms?

I am trying to research about speech recognition and why everyone uses Fourier transforms in going about the topic. I know that we get information related to the frequency of each sound uttered which ...
1
vote
1answer
107 views

Povey window formula

In Kaldi, "povey" is a window made to be similar to Hamming but to go to zero at the edges, it's: $$g(n) = \left(\frac{1-\cos\left(\frac{n}{N2\pi}\right)}{2}\right)^{0.85}$$ Please, can anyone ...
1
vote
1answer
64 views

How many psychoacoustical critical bands are there?

I understand that the human ear works as a band-pass filter and I have doubt regarding the number of such bandpass filters in the filter bank. I came across these two possible explanations of how it ...
1
vote
1answer
1k views

How to convert NIST to RIFF wav file

I have a dataset of audios(namely TIDIGITS, collected at Texas Instruments, Inc.) and they are in .wav format. However, when I try to read them from python, I get the following error: ...