We lifted 7 functions out of this paper's own repositories and ran 6 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| JoungheeKim/Non-Attentive-Tacotron | pwc_unofficial | 3 of 4 |
| IMDxD/NonAttentiveTacotron | pwc_unofficial | 3 of 3 |
| Function | Status | Where it lives |
|---|---|---|
| get_abspath | Ran | JoungheeKim/Non-Attentive-Tacotron/tacotron/utils.py code served (permissive licence) · get_code("cb38d9b9c8bab789") |
| get_mask_from_lengths | Ran | IMDxD/NonAttentiveTacotron/src/model/utils.py code served (permissive licence) · get_code("08d6a5b23b6ff882") |
| get_trellis | Ran | JoungheeKim/Non-Attentive-Tacotron/build_aligned_info.py code served (permissive licence) · get_code("ea26bcaea1557fe9") |
| load_filepaths_and_text | Ran | IMDxD/NonAttentiveTacotron/src/model/utils.py code served (permissive licence) · get_code("a79e0b1cd209a59b") |
| load_wav_to_torch | Ran | IMDxD/NonAttentiveTacotron/src/model/utils.py code served (permissive licence) · get_code("8ccc2e9689ef3ce4") |
| merge_repeats | Ran | JoungheeKim/Non-Attentive-Tacotron/build_aligned_info.py code served (permissive licence) · get_code("2ddca714d7ff5828") |
| backtrack | Not yet run | JoungheeKim/Non-Attentive-Tacotron/build_aligned_info.py code served (permissive licence) · get_code("499d76d1bd20ad71") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
This paper presents Non-Attentive Tacotron based on the Tacotron 2 text-to-speech model, replacing the attention mechanism with an explicit duration predictor. This improves robustness significantly as measured by unaligned duration ratio and word deletion rate, two metrics introduced in this paper for large-scale robustness evaluation using a pre-trained speech recognition model. With the use of Gaussian upsampling, Non-Attentive Tacotron achieves a 5-scale mean opinion score for naturalness of 4.41, slightly outperforming Tacotron 2. The duration predictor enables both utterance-wide and per-phoneme control of duration at inference time. When accurate target durations are scarce or unavailable in the training data, we propose a method using a fine-grained variational auto-encoder to train the duration predictor in a semi-supervised or unsupervised manner, with results almost as good as supervised training.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2010.04301")
get_code_for_paper("2010.04301")
have("2010.04301")
Connect an agent — have() is free.