69 Hits in 5.1 sec

Disentangling Timbre and Singing Style with Multi-singer Singing Synthesis System [article]

Juheon Lee, Hyeong-Seok Choi, Junghyun Koo, Kyogu Lee
2019 arXiv   pre-print
In this study, we define the identity of the singer with two independent concepts - timbre and singing style - and propose a multi-singer singing synthesis system that can model them separately.  ...  timbre and singing style.  ...  PROPOSED SYSTEM We propose a multi-singer SVS system that can model timbre and singing styles independently.  ... 
arXiv:1910.13069v1 fatcat:6uujv4fg35d6lbligrmirzeqcy

U-Singer: Multi-Singer Singing Voice Synthesizer that Controls Emotional Intensity [article]

Sungjae Kim, Kihyun Na, Choonghyeon Lee, Jehyeon An, Injung Kim
2022 arXiv   pre-print
We propose U-Singer, the first multi-singer emotional singing voice synthesizer that expresses various levels of emotional intensity.  ...  The visualization of the unified embedding space exhibits that U-singer estimates the correct variations in pitch and energy highly correlated with the singer ID and emotional intensity level.  ...  On the other hand, [39] proposes a method to disentangle timbre from singing style and control them separately.  ... 
arXiv:2203.00931v1 fatcat:s7lk5lq27bdgnapqyp57saaa5a

Learn2Sing: Target Speaker Singing Voice Synthesis by learning from a Singing Teacher [article]

Heyang Xue, Shan Yang, Yi Lei, Lei Xie, Xiulin Li
2020 arXiv   pre-print
Singing voice synthesis has been paid rising attention with the rapid development of speech synthesis area.  ...  and style tag embedding.  ...  Note that DAT has been recently used to disentangle speakers in multi-speaker singing synthesis [17] .  ... 
arXiv:2011.08467v1 fatcat:lb75zqy7gjd7dek3nboajbmpbi

A Comprehensive Survey on Deep Music Generation: Multi-level Representations, Algorithms, Evaluations, and Future Directions [article]

Shulei Ji, Jing Luo, Xinyu Yang
2020 arXiv   pre-print
audio generation converts scores with performance characteristics into audio by assigning timbre or generates music in audio format directly.  ...  However, the development history, the model evolution, as well as the pros and cons of same music generation task have not been clearly illustrated.  ...  [229] also extended the above-mentioned single-singer song synthesis system into a multi-singer system.  ... 
arXiv:2011.06801v1 fatcat:cixou3d2jzertlcpb7kb5x5ery

Explorations of Singing Voice Synthesis using DDSP

Juan Alonso, Cumhur Erkut
2021 Zenodo  
Machine learning based singing voice models require large datasets and lengthy training times.  ...  Our results indicate that the latent-𝑧 improves both the identification of the singer as well as the comprehension of the lyrics.  ...  Acknowledgments The authors would like to thank the following singers for their generous collaboration providing the audio files: Servando Carballar from Avi-  ... 
doi:10.5281/zenodo.5043851 fatcat:qjz26k6j7vdmvjfujqnubg3s2a

Speech-to-Singing Conversion based on Boundary Equilibrium GAN [article]

Da-Yi Wu, Yi-Hsuan Yang
2020 arXiv   pre-print
This is achieved by viewing speech-to-singing conversion as a style transfer problem.  ...  Specifically, given a speech input, and optionally the F0 contour of the target singing, the proposed model generates as the output a singing signal with a progressive-growing encoder/decoder architecture  ...  Due to the small size of the NUS dataset, we perform data augmentation using unpaired data with the DAMP dataset [23] , a multi-singer, singing-only dataset comprising of performances of 5,690 unique  ... 
arXiv:2005.13835v3 fatcat:h3adv25m5fcctdzefu6qdxj7li

Speech-to-Singing Conversion Based on Boundary Equilibrium GAN

Da-Yi Wu, Yi-Hsuan Yang
2020 Interspeech 2020  
This is achieved by viewing speech-to-singing conversion as a style transfer problem.  ...  Specifically, given a speech input, and the F0 contour of the target singing output, the proposed model generates the spectrogram of a singing signal with a progressive-growing encoder/decoder architecture  ...  Due to the small size of the NUS dataset, we perform data augmentation using unpaired data with the DAMP dataset [23] , a multi-singer, singing-only dataset comprising of performances of 5,690 unique  ... 
doi:10.21437/interspeech.2020-1984 dblp:conf/interspeech/WuY20 fatcat:273mvs5hybh2zljoyvjyxmdxc4

Learn2Sing 2.0: Diffusion and Mutual Information-Based Target Speaker SVS by Learning from Singing Teacher [article]

Heyang Xue, Xinsheng Wang, Yongmao Zhang, Lei Xie, Pengcheng Zhu, Mengxiao Bi
2022 arXiv   pre-print
During the training, to avoid the information confusion of the speaker embedding and the style embedding, mutual information is employed to restrain the learning of speaker embedding and style embedding  ...  Inspired by the fact that pitch is the key style factor to distinguish singing from speaking voice, the proposed Learn2Sing 2.0 first generates the preliminary acoustic feature with averaged pitch value  ...  To be specific, the singing data recorded by the female singer contains 100 songs with around 5 hours, and each song is labeled with the music scores and the duration of each phoneme.  ... 
arXiv:2203.16408v2 fatcat:zjywlgtpkreclc5k3c5pdndg5u

Knowledge-Based Probabilistic Modeling For Tracking Lyrics In Music Audio Signals

Georgi Dzhambazov, Xavier Serra
2017 Zenodo  
This confirms that music-specific knowledge is an important stepping stone for computationally tracking lyrics, especially in the challenging case of singing with instrumental accompaniment.  ...  In this thesis, we devise computational models for tracking sung lyrics in multi-instrumental music recordings.  ...  We introduced their training procedure in Section 2.2.8 and will refer  ... 
doi:10.5281/zenodo.841980 fatcat:tohf6dcvobhe3ei77nvp3wg3ba

Jukebox: A Generative Model for Music [article]

Prafulla Dhariwal, Heewoo Jun, Christine Payne, Jong Wook Kim, Alec Radford, Ilya Sutskever
2020 arXiv   pre-print
We can condition on artist and genre to steer the musical and vocal style, and on unaligned lyrics to make the singing more controllable.  ...  We introduce Jukebox, a model that generates music with singing in the raw audio domain.  ...  They can capture melody, rhythm, long-range composition, and timbres for a wide variety of instruments, as well as the styles and voices of singers to be produced with the music.  ... 
arXiv:2005.00341v1 fatcat:drwspmscbjfknhqdlunbp6spkm

An Acoustical Study of Individual Voices in Choral Blend

Allen W. Goodwin
1980 Journal of research in music education  
One aspect of the study involved identifying and comparing the acoustical qualities of sounds produced in the usual solo manner and sounds produced by the same singers attempting to blend with a unison  ...  Thirty sopranos were involved singing [CL], [0], [L ], [ s], and [ L ] on the pitches C 4 (261 Hz.), A 4 (440 Hz.), and F 5 (698 Hz).  ...  One of the ensemble singers possessed a large, dark, "operatic style" voice. Another singer possessed a small, light, bright voice.  ... 
doi:10.1177/002242948002800205 fatcat:424akjqpnzampgfysaa2usquu4

Universals in the world's musics

Steven Brown, Joseph Jordania
2011 Psychology of Music  
Comparison makes possible the analysis and the exact description of an individual phenomenon by comparing it with other phenomena and by emphasizing its distinctive qualities.  ...  These universals span a wide variety of features, including pitch, rhythm, melodic structure, form, vocal style, expressive devices, instruments, performance contexts, contents, and behaviors.  ...  Acknowledgments We are grateful to Patrick Savage and to the students in Dr.  ... 
doi:10.1177/0305735611425896 fatcat:tpkzxcxd7vg7rpac34h2m3qjjm

Dagstuhl Reports, Volume 9, Issue 1, January 2019, Complete Issue [article]

This has led to many practical engineering applications based on computational tasks such as singing assessment (linked to pitch, timbre and timing description), voice separation, singing synthesis, singer  ...  More recently, singing-synthesis systems have become widely used, and people actively enjoy songs with synthesized singing voices as the main vocals.  ...  Furthermore, we equipped one singer with an ambulatory monitoring system to acquire behavioral and physiological data during the performances. 13 Another singer was equipped with binaural microphones to  ... 
doi:10.4230/dagrep.9.1 fatcat:m3grhk5hanccbg7oxkhos7kv4e

"Let's Listen with Our Eyes ..." The Deconstruction of Deafness in Christine Sun Kim's Sound Art [chapter]

Anna Benedikt
2021 Under Construction: Performing Critical Identity  
[CrossRef] Anderson, Leon. 2006 Almost all of the 'core members' of the group were white, with Latinos and blacks following (Neumann 2008, p. 59), perhaps solely due to the group's positioning in the  ...  This romanticized identification fails, however, to account for the specificities of race relations within the group, as is common to attempts to replace racial categories with national ones, and which  ...  the synthesis of the arts.  ... 
doi:10.3390/books978-3-03897-500-7-3 fatcat:q7ca7vkxxfh43lizxsvkamgxla

Legitimate Voices: A Multi-Case Study of Trans and Non-Binary Singers in the Applied Voice Studio

William R. Sauerland
This qualitative, multi-case study examined trans and non-binary singers in the applied voice studio.  ...  The purpose of this study was to explore (1) the impact of music participation on the identities of trans and non-binary singers, (2) the experiences of trans and non-binary singers taking private singing  ...  Analysis and Synthesis of Data The purpose of this interpretive multi-case study is to explore the experiences of four trans or non-binary singers and their teachers in the applied voice studio, and thus  ... 
doi:10.7916/d8c26czh fatcat:6qyuzrnlobfpldzzjklf32brz4
« Previous Showing results 1 — 15 out of 69 results