Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Sign up for free
Menu
Search
Features
All features
Private URLs
Password Protection
Custom URLS
Scheduled publishing
Remove Branding
Restrict embedding
Deck Collections
Notes
Features
All features
Private URLs
Password Protection
Custom URLS
Scheduled publishing
Remove Branding
Restrict embedding
Deck Collections
Notes
Explore
Featured decks
Featured speakers
Programming
Technology
Storyboards
Explore
Featured decks
Featured speakers
Programming
Technology
Storyboards
Pricing
Search
Sign in
Sign up for free
TTS Skins: Speaker Conversion via ASR
Search
peisuke
November 20, 2020
Technology
470
0
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
TTS Skins: Speaker Conversion via ASR
Interspeech2020音声読み会発表資料
peisuke
November 20, 2020
Transcript
14 slides
TTS Skins: Speaker Conversion via ASR Authors: A. Polyak, L.
Wolf, Y. Taigman presenter: @peisuke
2016 ABEJA 2016 Twitter @peisuke Github https://github.com/peisuke Qiita https://qiita.com/peisuke SlideShare
https://www.slideshare.net/FujimotoKeisuke
• • TTS Skins: Speaker Conversion via ASR • •
• ASR WaveNet • • ASR
• Text-to-Speech 100 • • Text-to-Speech • • TTS
• ASR F0
• • Jasper: An End-to-End Convolutional Neural Acoustic Model •
https://github.com/NVIDIA/OpenSeq2Seq • • 1DConv-BN-ReL • Skip-Connection • Pre-trained •
• WaveNet • condition • https://github.com/NVIDIA/nv-WaveNet • • • •
F0 •
• • Look up table pytorch Embedding • • •
F0 • • fine tuning
• • LibriTTS VCTK • • Many-to-many seen unseen •
TTS • • • MOS • Mel cepstral distortion • Speaker classification • • WaveNet AutoEncoder • PPG
Seen • Seen-to-seen • A B • • Identification F0
LibriTTS VCTK MOS MCD Identification MOS MCD Identification Full method 3.78±0.83 96.12 4.08±0.75 8.76±1.72 98.97 w/o F0 3.61±0.83 96.96 3.59±0.96 8.99±1.5 96.89 AE baseline 2.89±0.88 29.19 3.46±1.07 9.45±1.63 69.26 PPG 2.82±0.91 94.01 2.67±0.93 9.19±1.50 98.77 PPG2 2.87±1.00 95.77 3.03±1.06 9.18±1.52 96.24
Uneen • Uneen-to-seen • A B • LibriTTS VCTK MOS
MCD Identification MOS MCD Identification Full method 3.70±0.80 97.10 4.05±0.74 8.94±1.53 98.33 w/o F0 3.67±0.82 97.15 3.62±0.99 9.25±1.62 95.69 AE baseline 3.02±0.89 32.55 3.83±0.91 9.65±1.51 66.20 PPG 2.79±0.93 94.05 2.89±0.93 9.45±1.45 97.45 PPG2 2.71±0.93 95.43 3.19±1.04 9.79±1.86 97.25
TTS • TTS • TTS LibriTTS VCTK MOS MCD Identification
MOS MCD Identification Original TTS 4.25±0.77 10.12±1.27 4.37±0.80 14.52±2.40 Full method 3.67±0.81 8.13±0.95 96.06 4.17±0.88 12.68±2.17 99.25 w/o F0 3.47±0.76 8.43±0.97 96.66 3.75±1.07 13.06±2.26 96.36 AE baseline 3.02±0.84 9.38±1.09 60.26 3.85±1.05 13.81±2.29 75.56 PPG 2.91±0.94 8.52±0.93 96.63 3.50±0.83 12.45±1.92 98.36 PPG2 2.85±0.87 8.76±1.06 95.08 3.66±1.03 12.57±2.10 97.62
• The voice conversion challenge 2018 • 1 81 4-5
• Hub Spoke • Hub Spoke MOS Similarity MOS Similarity Ours 3.84±0.85 2.87±1.14 4.00±0.55 3.14±0.97 N10 3.92±0.75 2.83±1.20 3.98±0.52 3.13±0.97 N17 3.27±0.95 2.77±1.17 3.40±0.88 3.05±0.96
• • TTS • ASR F0 Conditional WaveNet •
More Decks by peisuke
See All by peisuke
EgoX: Egocentric Video Generation from a Single Exocentric Video
peisuke
0
220
Moto: Latent Motion Token as the Bridging Language for Learning Robot Manipulation from Videos
peisuke
0
320
VGGT: Visual Geometry Grounded Transformer
peisuke
1
2.2k
AI for Kids:小学生に画像認識を教えてみた話
peisuke
1
130
LangGraphで始めるマルチエージェントシステム
peisuke
14
5.2k
Self-RAG: Learning to Retrieve, Generate and Critique through Self-Reflections
peisuke
9
1.7k
Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fields
peisuke
0
14k
LangChain Toolsの運用と改善
peisuke
5
3k
GNeRF: GAN-based Neural Radiance Field without Posed Camera
peisuke
1
880
Other Decks in Technology
See All in Technology
81天的 AI Agent 軟體開發團隊治理實務分享
regionbbs
0
180
[関西Kaggler会 20261113] 続・コミュニティコンペのススメ
nejumi
0
110
おうち環境でも「なるべく止めたくない!」 ― おうち環境の機器冗長化の理想と現実
yuutosi
1
500
農家は秘伝のタレを知らない - NOBODY KNOWS THE HIDDEN SOURCE - Kaggriculture 3rd Place Solution
masspeaks
2
480
BALLAS Technology Deck
ballas_inc
0
780
Quick Automate | 目指せ Quick マスター 12
inariku
PRO
1
120
Skill とカスタムエージェント (手順と役割を渡す) | 目指せ Kiro マスター 08
inariku
PRO
2
110
LLM 基盤と kagent v0 を全部 k8s に入れてみた / Deploy both LLM and kagent v0 in k8s
kota2and3kan
0
230
Sonnetだけで業務はまわる?
kintotechdev
1
140
専門家に使ってもらうAIプロダクトをつくる - 病院の医事課向けDPCコーディング支援システムの設計と評価
hagino3000
2
140
お風呂場をサーバルームにしよう
yusuke427
0
280
小さなエージェントから始めてみよう!実践コンテキストエンジニアリング
iwamot
PRO
0
130
Featured
See All Featured
How STYLIGHT went responsive
nonsquared
100
6.4k
The Art of Programming - Codeland 2020
erikaheidi
57
14k
Odyssey Design
rkendrick25
PRO
3
950
RailsConf 2023
tenderlove
30
1.6k
Building a Modern Day E-commerce SEO Strategy
aleyda
45
9.3k
So, you think you're a good person
axbom
PRO
2
2.3k
Sam Torres - BigQuery for SEOs
techseoconnect
PRO
0
610
The Cost Of JavaScript in 2023
addyosmani
56
10k
Darren the Foodie - Storyboard
khoart
PRO
5
4.1k
JavaScript: Past, Present, and Future - NDC Porto 2020
reverentgeek
52
6.2k
Navigating Team Friction
lara
192
16k
Balancing Empowerment & Direction
lara
6
1.4k