Bernardo Torres (@torres_be_) profile picture

Researcher

Joined January 2023
Photo by Bernardo Torres (@torres_be_)Photo by Bernardo Torres (@torres_be_)Photo by Bernardo Torres (@torres_be_)Photo by Bernardo Torres (@torres_be_)Photo by Bernardo Torres (@torres_be_)
Bernardo Torres retweeted
We just released the full training code, as well as our best pretrained model! 🎉 Feel free to use our SOTA checkpoint in your own project with 3 lines of code, or to retrain on your own data using our Lightning+Hydra+Dora codebase ⚡️🐍 🌐 github.com/sony/sampleid
Eminem sampled Aerosmith, 50 Cent sampled Nina Simone, everybody sampled Chic... Many great songs sampled existing ones! Detecting this is the topic of our latest paper with @serrjoa at @SonyAI Barcelona 😎 tl;dr: multi-track dataset + few tricks = +18% boost over SOTA 🚀 1/N
1
5
1
17
1,807
The PESTO extension paper was published in TISMIR! Here's the link to the publication: transactions.ismir.net/artic… It features several improvements to the model, larger cross dataset evaluation, and a real-time implementation.
Estremamente felice di annunciare that our new PESTO recipe has been published in the famous TISMIR cookbook 👨‍🍳🤌 With chef @torres_be_, we revisited this traditional sauce invented at ISMIR 2023 in Milan with some Brazilian flavours 🇮🇹🇧🇷 transactions.ismir.net/artic…
1
1
161
+ if you work on MIR and use the CQT for tasks which require frame-wise/time-varying estimation, consider using the Variable-Q Transform (VQT) instead. The huge CQT kernels at low frequencies are often underestimated and they might be hurting your performances.
19
Bernardo Torres retweeted
🔉New paper out! Recent audio codecs typically learn compression and quantization jointly, limiting the choice of quantization layers, non-differentiable by definition. What if we used powerful neural quantizers like Qinco2 and trained them offline? arxiv.org/abs/2503.19597
1
14
62
4,466
We’ve added fresh ingredients and cooking tricks to make the best, lightest, and fastest neural pitch estimator even better! 🌿🔥 Shoutout to all collaborators and to the amazing chef & SSL titan, @howariou
PESTO 2.0 è rilasciato! 🥳🥳🥳 With Brazilian chef @torres_be_ (and others), we revisit this traditional italian sauce, invented in Milan at @ISMIRConf 2023 🇮🇹 And you can taste it in REAL-TIME at home (~5 ms latency) ⏱️ 1/6
1
1
11
397
Bernardo Torres retweeted
PESTO 2.0 è rilasciato! 🥳🥳🥳 With Brazilian chef @torres_be_ (and others), we revisit this traditional italian sauce, invented in Milan at @ISMIRConf 2023 🇮🇹 And you can taste it in REAL-TIME at home (~5 ms latency) ⏱️ 1/6
1
7
2
24
2,234
🌟My keynote at the @c4dm workshop about "Models of Musical Signals: Representation, Learning & Generation" is now on YouTube, giving an overview on developments in self-supervised learning for audio since 2020, low-level representation learning, audio (stem) generation and much more 🧵👇 youtube.com/watch?v=ixHfBPXS… @SonyCSLMusic @SonyCSLParis
1
11
36
2,707
Interested in ill-posed learning tasks, uncertainty prediction, conditional density estimation or multi-head deep neural networks ? In our new paper, accepted at #ICML24, we tackle these challenges by exploring the Winner-Takes-All (WTA) training scheme. [1/n]
1
10
2
35
5,592
Signals with a high degree of autocorrelation (such as pitched signals) make the training of convnets on raw audio unstable. For Gaussian initialization, the greater the input’s autocorrelation, the greater the variance of the output. Huge relief seeing these kinds of papers <3
Training convnets on waveforms is hard—far harder than on magnitude spectrograms. "Instabilities in Convnets for Raw Audio" approaches this phenomenon from the perspective of sensitivity to initialization. IEEE Signal Processing Letters vol. 31 preprint: hal.science/hal-04528116
5
30
2,659
So many amazing people in this photo!
Replying to @AdaSp
@adasp group from ⁦⁦@telecomparis⁩ @ieeeICASSP⁩ 2014, Seoul, Korea
6
276
Happy to share some work accepted to ICASSP :) We experiment with a loss function inspired by optimal transport to compare spectra. We test it on a synthesis-based frequency localization (and F0 estimation) toy task using a harmonic synthesizer. Paper: arxiv.org/abs/2312.14507
10 papers from the Audio group @tp_adasp of Télécom-Paris will be presented at @icassp2014 @ieeeICASSP. --- Paper #1: Bernardo Torres et al. "Unsupervised Harmonic Parameter Estimation Using Differentiable DSP and Spectral Optimal Transport"
2
3
1
27
4,945
We call this loss Spectral Optimal Transport (SOT). Compared to L1 and L2 spectral losses on sinusoidal signals, SOT has a very nice convex curve leading to the right oscillator frequency. Its gradient also does not vanish when the frequency difference is high.
1
1
226
The SOT loss is based on the p-Wasserstein distance, which has a closed-form solution in 1D. First we compute the cumulative function of both spectra. Then, we invert it to get the quantiles. We add up the differences (raised to power p) and that’s it!
146
Bernardo Torres retweeted
🥳 We present our #ICASSP2024 paper: A diffusion model that generates production-quality (bass) audio stems to any audio input. 😎🎸 According to our experience, that's more useful to artists than generating full mixes. 🙃 📜Paper: arxiv.org/abs/2402.01412 🎶Demo: tinyurl.com/bass-acc-demo by @marco_ppasini 👈💪@SonyCSLMusic #MusicAI
3
24
7
112
20,290
Bernardo Torres retweeted
the math in most diffusion papers
33
74
30
845
115,123
Happy to be participating in this year’s #ISMIR2023!
3
322
Bernardo Torres retweeted
Huan had her visa to travel to Italy for @ISMIRConf rejected - for the second time in a row, despite having papers to present both years, which is incredibly unfair! If anyone can help, please get in touch 🙏
Does any one know lawyer in Italy and could reference me to? For visa refusals, the embassy told me to find lawyers in Italy. I would like to pay whatever I can to get it sorted…
15
1
23
5,916
Bernardo Torres retweeted
Oldies but goldies: H. Kuhn, The Hungarian Method for the assignment problem, 1955. The most celebrated algorithm to solve efficiently (almost cubic time) the assignment problem. en.wikipedia.org/wiki/Hungar…
5
126
4
613
65,737