{"url":"/method/gblock","slug":"gblock","name":"GBlock","full_name":"GBlock","full_name_withheld":false,"description_markdown":"**GBlock** is a type of [residual block](https://paperswithcode.com/method/residual-block) used in the [GAN-TTS](https://paperswithcode.com/method/gan-tts) text-to-speech architecture - it is a stack of two residual blocks. As the generator is producing raw audio (e.g. a 2s training clip corresponds\r\nto a sequence of 48000 samples), dilated convolutions are used to ensure that the receptive field of $G$ is large enough to capture long-term dependencies. The four kernel size-3 convolutions in each GBlock have increasing dilation factors: 1, 2, 4, 8. Convolutions are preceded by Conditional Batch Normalisation, conditioned on the linear embeddings of the noise term $z \\sim N\\left(0, \\mathbf{I}\\_{128}\\right)$ in the single-speaker case, or the concatenation of $z$ and a one-hot representation of the speaker ID in the multi-speaker case. The embeddings are different for\r\neach BatchNorm instance. \r\n\r\nA GBlock contains two skip connections, the first of which in [GAN](https://paperswithcode.com/method/gan)-TTS performs upsampling if the output frequency is higher than the input, and it also contains a size-1 [convolution](https://paperswithcode.com/method/convolution)\r\nif the number of output channels is different from the input.","description_state":"present","introduced_year":null,"introduced_by":{"title":"High Fidelity Speech Synthesis with Adversarial Networks","paper":"/paper/high-fidelity-speech-synthesis-with-1","first_author":"Mikołaj Bińkowski","n_authors":8,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/high-fidelity-speech-synthesis-with-1"},"source":{"url":"https://arxiv.org/abs/1909.11646v2","title":"High Fidelity Speech Synthesis with Adversarial Networks","url_on_a_paper_host":true},"code_snippet_url":"https://github.com/yanggeng1995/GAN-TTS/blob/4675fa108c4c52f190d27a32a8d9e9ce1c68d7a1/models/generator.py#L38","code_snippet_url_on_a_code_host":true,"categories":[{"area":"General","area_id":"general","collection":"Skip Connection Blocks","url":"/methods/category/skip-connection-blocks","pwc_aliases":[]},{"area":"Audio","area_id":"audio","collection":"Audio Model Blocks","url":"/methods/category/audio-model-blocks","pwc_aliases":[]}],"n_papers_tagged":2,"archive_num_papers":2,"papers_newest_first":[{"paper":"/paper/a-spectral-energy-distance-for-parallel","title":"A Spectral Energy Distance for Parallel Speech Synthesis","date":"2020-08-03","arxiv_id":"2008.01160","n_code_links":2,"syntology":{"ran":0,"of":3,"unverified":3,"pointer_only":0}},{"paper":"/paper/high-fidelity-speech-synthesis-with-1","title":"High Fidelity Speech Synthesis with Adversarial Networks","date":"2019-09-25","arxiv_id":"1909.11646","n_code_links":3,"syntology":{"ran":2,"of":5,"unverified":3,"pointer_only":4}}],"papers_shown":2,"tasks":[{"task":"/task/speech-synthesis","name":"Speech Synthesis","papers":2},{"task":null,"name":"Generative Adversarial Network","papers":1},{"task":"/task/text-to-speech","name":"Text to Speech","papers":1},{"task":"/task/high","name":"Vocal Bursts Intensity Prediction","papers":1},{"task":"/task/scoring-rule","name":"scoring rule","papers":1},{"task":"/task/text-to-speech-1","name":"text-to-speech","papers":1}],"tasks_shown":6,"n_tasks":6,"usage_by_year":[{"year":"2019","papers":1},{"year":"2020","papers":1}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/gblock"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}