An HMM-based singing voice synthesis system

By Keijiro Saino, Heiga Zen, Yoshihiko Nankaku, Akinobu Lee, Keiichi Tokuda. 2006

141 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Speech & Audio, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Xu Tan
2. Lior Wolf
3. Tao Qin
4. Junichi Yamagishi
5. Tie-Yan Liu
6. Zhou Zhao
7. Heiga Zen
8. Yi Ren

In Speech & Audio

1. Junichi Yamagishi
2. Lior Wolf
3. Xu Tan
4. Yi Ren
5. Yossi Adi
6. Eliya Nachmani
7. Zhou Zhao
8. Jinglin Liu

In Signal Processing

1. Lior Wolf
2. Xu Tan
3. Yossi Adi
4. Junichi Yamagishi
5. Yi Ren
6. Zhou Zhao
7. Heiga Zen
8. Jinglin Liu

Citations in the corpus

• Jukebox: A Generative Model for Music (2020, methodology): “…(Hiller Jr & Isaacson, 1957; Moorer, 1972; Hadjeres et al., 2017; Huang et al., 2017), digital vocoders generating a singer’s voice (Bonada & Serra, 2007; Saino et al., 2006; Blaauw & Bonada, 2017) and also synthesizers producing timbres for various musical instruments (Engel et al., 2017; 2019).”
• DiffSinger: Diffusion Acoustic Model for Singing Voice Synthesis (2021, methodology): “Initial works of singing voice synthesis generate the sounds using concatenated (Macon et al., 1997; Kenmochi & Ohshita, 2007) or HMM-based parametric (Saino et al., 2006; Oura et al., 2010) The voice naturalness and diversity of SVS system have been continuously improved in recent years.”
• Recent development of the HMM-based singing voice synthesis system - Sinsy (2010, methodology): “Figure 1 gives an overview of the HMM-based singing voice synthesis system [5].”
• Unsupervised Singing Voice Conversion (2019, methodology): “Singing Synthesis and Conversion Classical singing synthesis methods are mostly concatenative (unit selection) methods [13] or HMM based [14, 15].”
• A Survey on Recent Deep Learning-driven Singing Voice Synthesis Systems (2021, methodology): “Before neural networks were widely used, unit concatenation [28] and hidden Markov Model (HMM) [16] approaches were adopted for SVS.”
• XiaoiceSing: A High-Quality and Integrated Singing Voice Synthesis System (2020, methodology): “Traditional methods in SVS including unit concatenation [1, 2, 3, 4] and statistical parametric synthesis [5, 6] have been dominant for many years.”
• Speech Synthesis Based on Hidden Markov Models (2013, background): “Singing voice synthesis can be regarded as synthesis in a special language in which linguistic specifications and resulting contexts are derived from musical notes and lyrics [82], [83].”
• Unsupervised Cross-Domain Singing Voice Conversion (2020, methodology): “Singing Synthesis and Conversion Previous singing synthesis methods applied unit selection methods [15] or HMM based parametric methods [16–18].”
• A Neural Parametric Singing Synthesizer (2017, methodology): “While naive implementations of the autoregressive generation algorithm tend to be inefficient, using a smart algorithm we can greatly speed up the process and obtain a system that’s competitive in both speed and quality.”