Alert button
Picture for Shih-Lun Wu

Shih-Lun Wu

Alert button

Music ControlNet: Multiple Time-varying Controls for Music Generation

Nov 13, 2023
Shih-Lun Wu, Chris Donahue, Shinji Watanabe, Nicholas J. Bryan

Figure 1 for Music ControlNet: Multiple Time-varying Controls for Music Generation
Figure 2 for Music ControlNet: Multiple Time-varying Controls for Music Generation
Figure 3 for Music ControlNet: Multiple Time-varying Controls for Music Generation
Figure 4 for Music ControlNet: Multiple Time-varying Controls for Music Generation
Viaarxiv icon

Improving Audio Captioning Models with Fine-grained Audio Features, Text Embedding Supervision, and LLM Mix-up Augmentation

Sep 29, 2023
Shih-Lun Wu, Xuankai Chang, Gordon Wichern, Jee-weon Jung, François Germain, Jonathan Le Roux, Shinji Watanabe

Figure 1 for Improving Audio Captioning Models with Fine-grained Audio Features, Text Embedding Supervision, and LLM Mix-up Augmentation
Figure 2 for Improving Audio Captioning Models with Fine-grained Audio Features, Text Embedding Supervision, and LLM Mix-up Augmentation
Figure 3 for Improving Audio Captioning Models with Fine-grained Audio Features, Text Embedding Supervision, and LLM Mix-up Augmentation
Figure 4 for Improving Audio Captioning Models with Fine-grained Audio Features, Text Embedding Supervision, and LLM Mix-up Augmentation
Viaarxiv icon

Listener Model for the PhotoBook Referential Game with CLIPScores as Implicit Reference Chain

Jun 16, 2023
Shih-Lun Wu, Yi-Hui Chou, Liangze Li

Figure 1 for Listener Model for the PhotoBook Referential Game with CLIPScores as Implicit Reference Chain
Figure 2 for Listener Model for the PhotoBook Referential Game with CLIPScores as Implicit Reference Chain
Figure 3 for Listener Model for the PhotoBook Referential Game with CLIPScores as Implicit Reference Chain
Figure 4 for Listener Model for the PhotoBook Referential Game with CLIPScores as Implicit Reference Chain
Viaarxiv icon

Tensor decomposition for minimization of E2E SLU model toward on-device processing

Jun 02, 2023
Yosuke Kashiwagi, Siddhant Arora, Hayato Futami, Jessica Huynh, Shih-Lun Wu, Yifan Peng, Brian Yan, Emiru Tsunoo, Shinji Watanabe

Figure 1 for Tensor decomposition for minimization of E2E SLU model toward on-device processing
Figure 2 for Tensor decomposition for minimization of E2E SLU model toward on-device processing
Figure 3 for Tensor decomposition for minimization of E2E SLU model toward on-device processing
Figure 4 for Tensor decomposition for minimization of E2E SLU model toward on-device processing
Viaarxiv icon

The Pipeline System of ASR and NLU with MLM-based Data Augmentation toward STOP Low-resource Challenge

May 11, 2023
Hayato Futami, Jessica Huynh, Siddhant Arora, Shih-Lun Wu, Yosuke Kashiwagi, Yifan Peng, Brian Yan, Emiru Tsunoo, Shinji Watanabe

Figure 1 for The Pipeline System of ASR and NLU with MLM-based Data Augmentation toward STOP Low-resource Challenge
Figure 2 for The Pipeline System of ASR and NLU with MLM-based Data Augmentation toward STOP Low-resource Challenge
Figure 3 for The Pipeline System of ASR and NLU with MLM-based Data Augmentation toward STOP Low-resource Challenge
Figure 4 for The Pipeline System of ASR and NLU with MLM-based Data Augmentation toward STOP Low-resource Challenge
Viaarxiv icon

A Study on the Integration of Pipeline and E2E SLU systems for Spoken Semantic Parsing toward STOP Quality Challenge

May 06, 2023
Siddhant Arora, Hayato Futami, Shih-Lun Wu, Jessica Huynh, Yifan Peng, Yosuke Kashiwagi, Emiru Tsunoo, Brian Yan, Shinji Watanabe

Figure 1 for A Study on the Integration of Pipeline and E2E SLU systems for Spoken Semantic Parsing toward STOP Quality Challenge
Figure 2 for A Study on the Integration of Pipeline and E2E SLU systems for Spoken Semantic Parsing toward STOP Quality Challenge
Figure 3 for A Study on the Integration of Pipeline and E2E SLU systems for Spoken Semantic Parsing toward STOP Quality Challenge
Viaarxiv icon

Compose & Embellish: Well-Structured Piano Performance Generation via A Two-Stage Approach

Sep 17, 2022
Shih-Lun Wu, Yi-Hsuan Yang

Figure 1 for Compose & Embellish: Well-Structured Piano Performance Generation via A Two-Stage Approach
Figure 2 for Compose & Embellish: Well-Structured Piano Performance Generation via A Two-Stage Approach
Figure 3 for Compose & Embellish: Well-Structured Piano Performance Generation via A Two-Stage Approach
Figure 4 for Compose & Embellish: Well-Structured Piano Performance Generation via A Two-Stage Approach
Viaarxiv icon

Theme Transformer: Symbolic Music Generation with Theme-Conditioned Transformer

Nov 07, 2021
Yi-Jen Shih, Shih-Lun Wu, Frank Zalkow, Meinard Müller, Yi-Hsuan Yang

Figure 1 for Theme Transformer: Symbolic Music Generation with Theme-Conditioned Transformer
Figure 2 for Theme Transformer: Symbolic Music Generation with Theme-Conditioned Transformer
Figure 3 for Theme Transformer: Symbolic Music Generation with Theme-Conditioned Transformer
Figure 4 for Theme Transformer: Symbolic Music Generation with Theme-Conditioned Transformer
Viaarxiv icon