Papers › Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions

Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions

16 Dec 2017arXiv:1712.05884archive 2025-07-28

Jonathan Shen, Ruoming Pang, Ron J. Weiss, Mike Schuster, Navdeep Jaitly, Zongheng Yang, Zhifeng Chen, Yu Zhang, Yuxuan Wang, RJ Skerry-Ryan, Rif A. Saurous, Yannis Agiomyrgiannakis, Yonghui Wu

This paper describes Tacotron 2, a neural network architecture for speech synthesis directly from text. The system is composed of a recurrent sequence-to-sequence feature prediction network that maps character embeddings to mel-scale spectrograms, followed by a modified WaveNet model acting as a vocoder to synthesize timedomain waveforms from those spectrograms. Our model achieves a mean opinion score (MOS) of $4.53$ comparable to a MOS of $4.58$ for professionally recorded speech. To validate our design choices, we present ablation studies of key components of our system and evaluate the impact of using mel spectrograms as the input to WaveNet instead of linguistic, duration, and F₀ features. We further demonstrate that using a compact acoustic intermediate representation enables significant simplification of the WaveNet architecture.

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="1712.05884")

Code

Syntology Ran 7 of 7 code samples harvested from 5 repositories linked to this paper; 0 have no recorded run. Of those that ran: 2 ran · honoured contract; 5 ran · our draft was wrong.

By repository: community (archive-listed): 6 samples from 5 repositories, 6 ran; 1 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

33 repositories listed; official and paper-mentioned ones first.

BogiHsu/Tacotron2-PyTorch mentioned on GitHubpytorch report
Jeevesh8/Cross-Lingual-Voice-Cloning mentioned on GitHubpytorchBSD-3-Clause report
NVIDIA/tacotron2 mentioned on GitHubpytorchBSD-3-Clause report
OlaWod/my-tacotron2 mentioned on GitHubpytorch report
Rayhane-mamah/Tacotron-2 mentioned on GitHubtf report
TensorSpeech/TensorflowTTS mentioned on GitHubtf report
alpharol/Taco_Collection mentioned on GitHubtf report
anandaswarup/TTS mentioned on GitHubpytorchMIT report
anandaswarup/rnn-tts mentioned on GitHubpytorch report
bfs18/tacotron2 mentioned on GitHubpytorch report
choiHkk/Transformer-TTS mentioned on GitHubpytorch report
coqui-ai/TTS mentioned on GitHubpytorchMPL-2.0 report
creotiv/RussianTTS-Tacotron2 mentioned on GitHubpytorch report
dathudeptrai/TensorflowTTS mentioned on GitHubtf report
dipjyoti92/SC-WaveRNN mentioned on GitHubpytorch report
dipjyoti92/TTS-Style-Transfer mentioned on GitHubpytorchMIT report
izzajalandoni/tts_models mentioned on GitHubpytorchBSD-3-Clause report
kaiidams/voice100 mentioned on GitHubpytorch report
kaiidams/voice100-tts mentioned on GitHubpytorch report
keonlee9420/Comprehensive-Tacotron2 mentioned on GitHubpytorchMIT report
martinlenglet/avtacotron2 mentioned on GitHubpytorch report
rosinality/melgan-pytorch mentioned on GitHubpytorch report
s3nh/pytorch-tacotron2 mentioned on GitHubpytorch report
thepowerfuldeez/tacotron2 mentioned on GitHubpytorch report
thuhcsi/tacotron mentioned on GitHubpytorch report
vincenzo-scotti/tacotron2 mentioned on GitHubpytorch report
xcmyz/FastSpeech mentioned on GitHubpytorch report
xinshengwang/Tacotron-pytorch mentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

7 samples harvested; 7 ran; 2 honoured the contract we drafted; 0 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

2ran · honoured contract
5ran · our draft was wrong

Licence: 2 of the 7 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 5 repositories linked to this paper, official or community; each sample names its own and says which. Some samples are identical code Syntology first harvested from another repository; for those, this paper's copy is not located and its licence is not recorded. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

length_to_mask rosinality/melgan-pytorch/tacotron2.py community (archive-listed) ran · our draft was wrong MIT (permissive) · 02ce0b07ac0aa7e2 · report
load_checkpoint anandaswarup/rnn-tts/train_Tacotron2.py community (archive-listed) ran · honoured contract MIT (permissive) · 669fcedc80c2fee3 · report
load_trained_model anandaswarup/rnn-tts/tts_Synthesis.py community (archive-listed) ran · our draft was wrong MIT (permissive) · 91be03213654f639 · report
np_now dipjyoti92/SC-WaveRNN/train_tacotron.py community (archive-listed) ran · our draft was wrong no licence file found · pointer only · 2af6a131384db5c8 · report
run_shell_command martinlenglet/avtacotron2/do_syn.py community (archive-listed) ran · our draft was wrong BSD-3-Clause (permissive) · 33ee8256742bf42c · report
split_func Rayhane-mamah/Tacotron-2/tacotron/models/tacotron.py community (archive-listed) ran · our draft was wrong MIT (permissive) · b1491d3c2e709fdc · report
get_param_num identical code first harvested elsewhere ran · honoured contract licence of this copy not recorded · f1cba67eb6d6e4ad · report

Tasks

Speech Synthesis

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Speech Synthesis North American English Tacotron 2 Mean Opinion Score 4.526 #1 of 7 Archive leaderboard report
Speech Synthesis North American English WaveNet (Linguistic) Mean Opinion Score 4.341 #2 of 7 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Introduced by this paper: Tacotron 2

Batch NormalizationBiLSTMConvolutionDilated Causal ConvolutionDropoutExponential DecayLSTMLinear LayerLocation Sensitive AttentionMax PoolingMixture of Logistic DistributionsReLUResidual ConnectionTacotron 2Tanh ActivationWaveNetWeight DecayZoneout

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections