STSM-FiLM: A FiLM-Conditioned Neural Architecture for Time-Scale Modification of Speech
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866916987176747008 |
|---|---|
| author | Wisnu, Dyah A. M. G. Zezario, Ryandhimas E. Rini, Stefano Li, Fo-Rui Peng, Yan-Tsung Wang, Hsin-Min Tsao, Yu |
| author_facet | Wisnu, Dyah A. M. G. Zezario, Ryandhimas E. Rini, Stefano Li, Fo-Rui Peng, Yan-Tsung Wang, Hsin-Min Tsao, Yu |
| contents | Time-Scale Modification (TSM) of speech aims to alter the playback rate of audio without changing its pitch. While classical methods like Waveform Similarity-based Overlap-Add (WSOLA) provide strong baselines, they often introduce artifacts under non-stationary or extreme stretching conditions. We propose STSM-FILM - a fully neural architecture that incorporates Feature-Wise Linear Modulation (FiLM) to condition the model on a continuous speed factor. By supervising the network using WSOLA-generated outputs, STSM-FILM learns to mimic alignment and synthesis behaviors while benefiting from representations learned through deep learning. We explore four encoder-decoder variants: STFT-HiFiGAN, WavLM-HiFiGAN, Whisper-HiFiGAN, and EnCodec, and demonstrate that STSM-FILM is capable of producing perceptually consistent outputs across a wide range of time-scaling factors. Overall, our results demonstrate the potential of FiLM-based conditioning to improve the generalization and flexibility of neural TSM models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_02672 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | STSM-FiLM: A FiLM-Conditioned Neural Architecture for Time-Scale Modification of Speech Wisnu, Dyah A. M. G. Zezario, Ryandhimas E. Rini, Stefano Li, Fo-Rui Peng, Yan-Tsung Wang, Hsin-Min Tsao, Yu Audio and Speech Processing Sound Time-Scale Modification (TSM) of speech aims to alter the playback rate of audio without changing its pitch. While classical methods like Waveform Similarity-based Overlap-Add (WSOLA) provide strong baselines, they often introduce artifacts under non-stationary or extreme stretching conditions. We propose STSM-FILM - a fully neural architecture that incorporates Feature-Wise Linear Modulation (FiLM) to condition the model on a continuous speed factor. By supervising the network using WSOLA-generated outputs, STSM-FILM learns to mimic alignment and synthesis behaviors while benefiting from representations learned through deep learning. We explore four encoder-decoder variants: STFT-HiFiGAN, WavLM-HiFiGAN, Whisper-HiFiGAN, and EnCodec, and demonstrate that STSM-FILM is capable of producing perceptually consistent outputs across a wide range of time-scaling factors. Overall, our results demonstrate the potential of FiLM-based conditioning to improve the generalization and flexibility of neural TSM models. |
| title | STSM-FiLM: A FiLM-Conditioned Neural Architecture for Time-Scale Modification of Speech |
| topic | Audio and Speech Processing Sound |
| url | https://arxiv.org/abs/2510.02672 |