12
votes
How can I upsample 22 kHz speech audio recording to 44 kHz, maybe using AI?
What you want to do is called "Super Resolution" in the field of imaging.
This is an ill posed problem.
Hence in order to solve it you need some prior / model about your audio data.
For ...
- 387
9
votes
How can I upsample 22 kHz speech audio recording to 44 kHz, maybe using AI?
While I don't know about tools for upsampling using AI, I am challenging the assumption that the main problem with your sample is lost high frequncies due to resampling from 44kHz to 22kHz, so "...
- 191
6
votes
How does Siri recognize me saying "Hey Siri"?
"Ok Google" is described in many publications by Google
Automatic Gain Control and Multi-style Training for Robust Small-Footprint Keyword Spotting with Deep Neural Networks
Convolutional Neural ...
- 1,119
4
votes
Converting Google's text-to-speech voice to, say, Justin Bieber's voice
How we should go about from converting Google Text to Speech's m3 file to Justin Bieber's voice?
Ask Mr. Bieber to repeat what your MP3 says.
I think letting one speech synthesizer first speak ...
- 32.5k
4
votes
How can I upsample 22 kHz speech audio recording to 44 kHz, maybe using AI?
As pointed in some of the comments, there are several issues involved in your problem that you should consider to go for an optimal solution:
Quality of the record system. Features of the microphone, ...
- 311
3
votes
Accepted
Male and female voice spectrum
in my opinion and limited experience, you can't just compare spectra of male and female voices to identify which is which.
2 decades ago i worked on two products called PurePitch and PitchDoctor that ...
- 21.4k
3
votes
Accepted
Anonymize / Obfuscate speech when doing audio classification
If I only use the images of power spectrograms are those images obfuscated enough to make a reconstruction of the audio signal not feasible?
That depends on how you parameterize your spectrogram. In ...
- 32.5k
2
votes
FFmpeg audio filter pipeline for speech enhancement
The 1st mentioned tutorial https://medium.com/@jud.dagnall/dynamic-range-compression-for-audio-with-ffmpeg-and-compand-621fe2b1a892 is a great introduction and the filter documentation also adds a few ...
2
votes
FFmpeg audio filter pipeline for speech enhancement
I just came across one article related to the compand command that might be useful (just in case someone looking for help sees this article).
From the article:
To test this, and my understanding ...
- 21
2
votes
Accepted
CODECS used for VoIP calls
Let me put in an unconnected form things about audio codecs.
Audio can be music or speech (discluding ultrasound, sonar etc).
Music is wideband and requires high-fidelity. Speech has a lower ...
- 28.4k
2
votes
Accepted
Given MFCC values for some wav’s, how can I easily tell if all speakers are the same?
Given MFCC values for some wav’s, how can I easily tell if all speakers are the same?
In general, you would deal with this kind of problem using some form of Clustering.
The fundamental premise ...
- 10.7k
2
votes
Accepted
How do I deepen my voice (I'm a female) to sound like a man in a video for a school project?
I know some audio software that does a good job:
Audacity (free software) - effect -> change pitch (set the pitch to down, lower down the semi-tones and the percent change)
Adobe Audition (paid ...
- 369
2
votes
make a voice sound signal play faster
You can increase the sampling rate, instead of the signal $x(nT_s) = x(n/F_s)$, you can play it at say $\frac{11}{10}F_s, \frac{12}{10}F_s, \text{or}\ 2F_s$. So you have:
\begin{align}
x\left(\frac{n}{...
- 3,426
2
votes
Obtain Envelope and Pitch curves of my Singing Voice
Why would you want to control an analog synth using your voice? Presumably because you want the «sound» of a synth patch coupled with the expressibility of your voice. Or only for the fun of making it ...
- 3,595
2
votes
Accepted
Speech recognition algorithm in GNU Octave to convert speech to text (and vice versa)
Is there any possibility to implement a speech recognition integrated with google voice that converts speechs in text in GNU Octave?
Yes.
The Google Cloud command line utility returns results in JSON....
- 10.7k
2
votes
How can I upsample 22 kHz speech audio recording to 44 kHz, maybe using AI?
Add high frequency noise.
While the other answers provide extremely valid points, adding noise is a very simple "solution" (deliberately placed in quotes) to make the sound appear more "...
- 121
2
votes
Accepted
Audio Processing - How to create a spectral pitch display?
You can usually trade off time versus frequency resolution in a spectrogram by changing the length, offset, and overlap of the STFT/FFTs.
If you compute more than one set of STFT/FFTs for a single ...
- 35.7k
2
votes
Accepted
Did Dialup Modem use closed or open loop power/volume control? How did they determine Tx level?
Did Dialup Modem use closed or open loop power/volume control?
Yes.
How did they determine Tx level?
"Dialup Modem" is a very big term that spans > 50 years of technological ...
- 32.5k
2
votes
Accepted
How to generate human voice fft?
generate human voice using fft
Can't do. The FFT is too primitive for anything but simplest audio generation: different voices, instruments, etc. will all have similar or identical-looking FFT. It's ...
- 9,242
1
vote
Meaning of frequency and bandwidth of a signal, despite the fact that we do not know the signal
I'm going to take a shot at this because I think your confusion is about how we use the word "bandwidth" and not necessarily a theoretical issue.
When we say a signal has bandwidth, it is ...
- 2,936
1
vote
Meaning of frequency and bandwidth of a signal, despite the fact that we do not know the signal
Some signals are generated by processes that are physically capable of generating frequencies only in a specific frequency range. For example, the human vocal tract can only generate signals between ...
- 15.4k
1
vote
Echo removal from a chunk of PCM signal
Your $S_{new}=N-F$ requires that you know both $N$ and $F$.
But since $N=F*R$, with $*$ being convolution, and $R$ being the Room Impulse Response (RIR), you're solving the equivalent problem $S_{new} ...
- 32.5k
1
vote
Accepted
what are the vector formats similar to hash codes to store user's voice for future voice verification?
Since you are not trying to verify if two "messages" are identical in a word-to-word sense, it would not be practical to go down the hash-function way.
This sounds more like a "pattern recognition" ...
- 10.7k
1
vote
Digital Audio Mixing Fundamental. What will be the data rate of mixed audio?
When you add multiple audio sources, you must first make sure that they use the same sampling rate, and then same numerical format, so that addition can be realized.
So, if they are not at the same ...
- 28.4k
1
vote
Adding White Gaussian noise to a voice signal
Your problem is how you calculate the signal power. You are calling norm which calculates $\sqrt{\sum_i |x_i|^2}$ but you want to calculate the sum of the squared ...
- 3,042
1
vote
Distort voice to hide talker identity as much as possible
To preserve the talker's identity for privacy reasons, I would make available for listening a distorted version of the original audio file, while using the original one for real recognition to improve ...
- 32.5k
1
vote
How does a telephone change your voice?
In the case of switched analog POTS phone lines, the audio response is shaped by the limited frequency response of the carbon microphone and the speaker in the handset.
- 35.7k
1
vote
Mathematical reason for MCPP in audio recognition?
The Mel is an empirical model that describes the change in human pitch perception at different frequencies: until 1 kHz we perceive pitch changes in a linear way, after that, we perceive it in a ...
1
vote
How to correctly classify voice from other sound events?
It is very possible that you don't have enough data to train your classifier. Although several thousand seems like a lot, it means that you only have a fraction of the actual thing you are trying to ...
- 101
1
vote
Accepted
Voice masking for removing voice and not birdsong
Sound masking in time-frequency is a very complex problem, especially to do it in real-time while collecting data. For ensuring privacy I would suggest using a simpler approach.
Sparse random ...
- 287
Only top scored, non community-wiki answers of a minimum length are eligible