SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2005.11129 · 2020

Glow-TTS: A Generative Flow for Text-to-Speech via Monotonic Alignment Search

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 14 functions out of this paper's own repositories and ran 8 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.

RepositoryRoleRan
ankurdhuriya/multispeaker-glow-tts pwc_unofficial 7 of 13
revsic/tf-glow-tts pwc_unofficial 1 of 1
FunctionStatusWhere it lives
basic_cleaners Ran ankurdhuriya/multispeaker-glow-tts/text/cleaners.py
code served (permissive licence) · get_code("a3a74a14c1d23fbb")
collapse_whitespace Ran ankurdhuriya/multispeaker-glow-tts/text/cleaners.py
code served (permissive licence) · get_code("7669c05d620e883e")
dynamic_range_compression Ran ankurdhuriya/multispeaker-glow-tts/audio_processing.py
code served (permissive licence) · get_code("83e2c862f7374ac7")
griffin_lim Ran ankurdhuriya/multispeaker-glow-tts/audio_processing.py
code served (permissive licence) · get_code("9f9ec26d4cdfdf7d")
latest_checkpoint_path Ran ankurdhuriya/multispeaker-glow-tts/utils.py
code served (permissive licence) · get_code("321604dda57bb02d")
load_state Ran revsic/tf-glow-tts/config.py
code served (permissive licence) · get_code("a93e65374343828b")
mask_from_len Ran ankurdhuriya/multispeaker-glow-tts/monotonic_align/monotonic_align/mas.py
code served (permissive licence) · get_code("f7da3590296389c4")
mask_from_lens Ran ankurdhuriya/multispeaker-glow-tts/monotonic_align/monotonic_align/mas.py
code served (permissive licence) · get_code("af9a236f8ff4a594")
duration_loss Not yet run ankurdhuriya/multispeaker-glow-tts/commons.py
code served (permissive licence) · get_code("2b6c52ad9b61c287")
intersperse Not yet run ankurdhuriya/multispeaker-glow-tts/commons.py
code served (permissive licence) · get_code("c01a3efb274ff778")
load_checkpoint Not yet run ankurdhuriya/multispeaker-glow-tts/utils.py
code served (permissive licence) · get_code("a4324b03a174abb9")
lowercase Not yet run ankurdhuriya/multispeaker-glow-tts/text/cleaners.py
code served (permissive licence) · get_code("296274b940cb0b32")
mle_loss Not yet run ankurdhuriya/multispeaker-glow-tts/commons.py
code served (permissive licence) · get_code("77759e4586956310")
plot_spectrogram_to_numpy Not yet run ankurdhuriya/multispeaker-glow-tts/utils.py
code served (permissive licence) · get_code("9dc202f235585e84")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Recently, text-to-speech (TTS) models such as FastSpeech and ParaNet have been proposed to generate mel-spectrograms from text in parallel. Despite the advantage, the parallel TTS models cannot be trained without guidance from autoregressive TTS models as their external aligners. In this work, we propose Glow-TTS, a flow-based generative model for parallel TTS that does not require any external aligner. By combining the properties of flows and dynamic programming, the proposed model searches for the most probable monotonic alignment between text and the latent representation of speech on its own. We demonstrate that enforcing hard monotonic alignments enables robust TTS, which generalizes to long utterances, and employing generative flows enables fast, diverse, and controllable speech synthesis. Glow-TTS obtains an order-of-magnitude speed-up over the autoregressive model, Tacotron 2, at synthesis with comparable speech quality. We further show that our model can be easily extended to a multi-speaker setting.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2005.11129")
get_code_for_paper("2005.11129")
have("2005.11129")

Connect an agent — have() is free.