profile
๐‘ฏ๐’๐’๐’†๐’”๐’•๐’š ๐‘ฐ๐’๐’•๐’†๐’ˆ๐’“๐’Š๐’•๐’š ๐‘ฌ๐’™๐’„๐’†๐’๐’๐’†๐’๐’„๐’†

Emotion Intensity and its Control for Emotional Voice Conversion

https://arxiv.org/abs/2201.03967?context=cs https://kunzhou9646.github.io/Emovox_demo/ Goal To explicitly characterize and control the intensity of emotion in voice conversion system Motivation Few ...

2022๋…„ 7์›” 13์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท

SpeechEQ: Speech Emotion Recognition based on Multi-scale Unified Datasets and Multitask Learning

Speech Emotion Recognition, Multi-task Learning,

2022๋…„ 7์›” 4์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท

Flow based model

Advatages of flow based model Exact latent-variable inference and log-likelihood evaluation --> latent variable์„ inferenceํ•  ๋•Œ ์ •ํ™•ํ•˜๊ฒŒ ํ•  ์ˆ˜ ์žˆ์Œ. original GAN, VAE์™€ ๋น„๊ตํ–ˆ์„ ๋•Œ, ๊ฑฐ์˜ exactlyํ•œ variable์„ ํ•™์Šตํ•  ์ˆ˜ ์žˆ์Œ. ์ •...

2022๋…„ 6์›” 14์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท

VScode extension

wavํŒŒ์ผ์„ vscode์—์„œ ๋ฐ”๋กœ ์žฌ์ƒํ•  ์ˆ˜ ์žˆ๋Š” extensionใ…Žใ… ์ง„์งœ ์‚ฌ๋žŒ๋“ค ์™œ์ผ€ ๋˜‘๋˜‘ใ…Ž ใ… ๋•ก์“ฐ a lot \~\~~ > \_ <

2022๋…„ 5์›” 26์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท

Generalized End-to-End Loss for Speaker Verification

https://arxiv.org/abs/1710.10467

2022๋…„ 5์›” 24์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท

Anaconda error: Solving environment: failed with initial frozen solve. Retrying with flexible solve

2022๋…„ 5์›” 19์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท

[LeetCode] 42. Trapping Rain Water

https://leetcode.com/problems/trapping-rain-water/ ๋…ผ๋ฌธ์ด ์ฝ๊ธฐ ์‹ซ์„ ๋•Œ์ฏค ํ‘ธ๋Š” ์•Œ๊ณ ๋ฆฌ์ฆ˜ ๋ฐฑ์ค€๋งŒ ํ’€๋‹ค๊ฐ€ ๋ฆฌํŠธ์ฝ”๋“œ๋กœ ๋„˜์–ด์™”๋‹ค. Level: Hard Problem Given n non-negative integers repres

2022๋…„ 5์›” 13์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท

Neural Voice Cloning with a Few Samples ๋…ผ๋ฌธ ์ •๋ฆฌ

#DL #speech #paper

2022๋…„ 5์›” 11์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท

Install selenium for mac

https://chromedriver.chromium.org/downloadspip3 install seleniumbrew install chromedriverbrew install --cask chromedriver

2022๋…„ 5์›” 9์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท

AUTOVC: Zero-Shot Voice Style Transfer with Only Autoencoder Loss

#DL #paper #speech

2022๋…„ 4์›” 22์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท

4์›”์˜ ๊ธฐ๋ก

-

2022๋…„ 4์›” 20์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท

ํฌ๋กฌ์—์„œ ํ”„๋กœ๊ทธ๋žจ ์„ค์น˜ ์—†์ด ๋™์˜์ƒ ๋ฐฐ์†ํ•˜๊ธฐ

์œˆ๋„์šฐ๋Š” F12, ๋งฅ์€ command + option + j ๋ˆŒ๋Ÿฌ์„œ ํฌ๋กฌ ๊ฐœ๋ฐœ์ž ๋„๊ตฌ๋ฅผ ์—ด์–ด์ค€๋‹ค. Console ์ฐฝ์— ์•„๋ž˜ ์ฝ”๋“œ๋ฅผ ์ž…๋ ฅํ•ด์ค€๋‹ค. 16๋ฐฐ๊ฐ€ ์ตœ๋Œ€์—ฌ์„œ, 17๋ถ€ํ„ฐ๋Š” ์—๋Ÿฌ๊ฐ€ ๋‚œ๋‹ค.

2022๋…„ 4์›” 18์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท

HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis ๋…ผ๋ฌธ ์ •๋ฆฌ

#Speech #DeepLearning #Paper

2022๋…„ 4์›” 10์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท

MelGAN: Generative Adversarial Networks for Conditional Waveform Synthesis ๋…ผ๋ฌธ ์ •๋ฆฌ

#Speech #DeepLearning #Paper

2022๋…„ 4์›” 7์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท

Colab ๋Ÿฐํƒ€์ž„ ์œ ์ง€

๋š๋š ๋Š๊ธฐ๋Š” ๋Ÿฐํƒ€์ž„๋งŒํผ ๐Ÿ‘‘๋ฐ›๋Š” ๊ฒƒ๋„ ์—†์ง€ Mac์€ command + option + j, Window๋Š” F12๋กœ ๊ฐœ๋ฐœ์ž ๋„๊ตฌ ์—ด์–ด์„œ ์ฝ˜์†” ์ฐฝ์— ์ž…๋ ฅํ•ด์ค์‹œ๋‹ค!

2022๋…„ 4์›” 7์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท

Parallel WaveNet: Fast High-Fidelity Speech Synthesis ๋…ผ๋ฌธ ์ •๋ฆฌ

#Speech #DeepLearning #Paper

2022๋…„ 4์›” 1์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท

WaveNet: A Generative Model for Raw Audio ๋…ผ๋ฌธ ์ •๋ฆฌ

#Speech #DeepLearning #Paper

2022๋…„ 3์›” 31์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท

Style Tokens: Unsupervised Style Modeling, Control and Transfer in End-to-End Speech Synthesis ๋…ผ๋ฌธ ์ •๋ฆฌ

https://arxiv.org/abs/1803.09017 Y. Wang et al., โ€œStyle Tokens: Unsupervised Style Modeling, Control and Transfer in End-to-End Speech Synthesis,โ€ in ICML, 2018. Summary E2E TTS์ธ ํƒ€์ฝ”ํŠธ๋ก  ๋‚ด์—์„œ ํ›ˆ๋ จ๋œ embeddi...

2022๋…„ 3์›” 30์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท

Prosody Tacotron: Towards End-to-End Prosody Transfer for Expressive Speech Synthesis with Tacotron ๋…ผ๋ฌธ ์ •๋ฆฌ

#Speech #DeepLearning #Paper

2022๋…„ 3์›” 29์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท

Deep Voice 2: Multi-Speaker Neural Text-to-Speech ๋…ผ๋ฌธ ์ •๋ฆฌ

์•„์ง ์ •๋ฆฌ ์•ˆํ–ˆ์ง€๋กฑ .. .

2022๋…„ 3์›” 29์ผ
ยท
0๊ฐœ์˜ ๋Œ“๊ธ€
ยท