Papers › FastSpeech 2: Fast and High-Quality End-to-End Text to Speech

FastSpeech 2: Fast and High-Quality End-to-End Text to Speech

8 Jun 2020ICLR 2021 1arXiv:2006.04558archive 2025-07-28

Yi Ren, Chenxu Hu, Xu Tan, Tao Qin, Sheng Zhao, Zhou Zhao, Tie-Yan Liu

Non-autoregressive text to speech (TTS) models such as FastSpeech can synthesize speech significantly faster than previous autoregressive models with comparable quality. The training of FastSpeech model relies on an autoregressive teacher model for duration prediction (to provide more information as input) and knowledge distillation (to simplify the data distribution in output), which can ease the one-to-many mapping problem (i.e., multiple speech variations correspond to the same text) in TTS. However, FastSpeech has several disadvantages: 1) the teacher-student distillation pipeline is complicated and time-consuming, 2) the duration extracted from the teacher model is not accurate enough, and the target mel-spectrograms distilled from teacher model suffer from information loss due to data simplification, both of which limit the voice quality. In this paper, we propose FastSpeech 2, which addresses the issues in FastSpeech and better solves the one-to-many mapping problem in TTS by 1) directly training the model with ground-truth target instead of the simplified output from teacher, and 2) introducing more variation information of speech (e.g., pitch, energy and more accurate duration) as conditional inputs. Specifically, we extract duration, pitch and energy from speech waveform and directly take them as conditional inputs in training and use predicted values in inference. We further design FastSpeech 2s, which is the first attempt to directly generate speech waveform from text in parallel, enjoying the benefit of fully end-to-end inference. Experimental results show that 1) FastSpeech 2 achieves a 3x training speed-up over FastSpeech, and FastSpeech 2s enjoys even faster inference speed; 2) FastSpeech 2 and 2s outperform FastSpeech in voice quality, and FastSpeech 2 can even surpass autoregressive models. Audio samples are available at https://speechresearch.github.io/fastspeech2/.

PaperPDFConference PDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 119 code samples harvested from 16 repositories linked to this paper: 83 ran (of which 27 constructed an object rather than computing a result; 70 with no instrument failure: 9 honoured, 1 violated, 60 with no contract checked; 13 where Syntology's instrument failed) · 36 unverified. Where Syntology's instrument failed: 12 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 119 samples from 16 repositories, 83 ran (70 with no instrument failure; 13 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 13 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 83 samples that ran has no recorded date.

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

KevinMIN95/StyleSpeech mentioned on GitHubpytorchSyntology: 8 samples, 6 ran (1 instrument-failed) · 2 unverified report
Munna-Manoj/Team6_FastSpeech2_TTS mentioned on GitHubpytorchSyntology: 2 samples, 0 ran · 2 unverified report
OlaWod/my-fastspeech2 mentioned on GitHubpytorchSyntology: 11 samples, 5 ran (0 instrument-failed) · 6 unverified report
RayeRen/RayeRen mentioned on GitHubSyntology: not harvested report
TensorSpeech/TensorflowTTS mentioned on GitHubtfSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis (arXiv:2010.05646; that paper's own run record: 17 ran (of which 11 constructed an object rather than computing a result; 14 with no instrument failure: 0 honoured, 0 violated, 14 with no contract checked; 3 where Syntology's instrument failed) · 8 unverified); nothing checks that they implement this paper's method. report
ai-unicamp/tts-objective-metrics mentioned on GitHubpytorchSyntology: not harvested report
as-ideas/TransformerTTS mentioned on GitHubtfNOASSERTIONSyntology: not harvested report
cadia-lvl/fastspeech2 mentioned on GitHubpytorchMITSyntology: 2 samples, 1 ran (1 instrument-failed) · 1 unverified report
coqui-ai/TTS mentioned on GitHubpytorchMPL-2.0Syntology: not harvested report
dathudeptrai/TensorflowTTS mentioned on GitHubtfSyntology: 21 samples, 15 ran (1 instrument-failed) · 6 unverified report
ga642381/FastSpeech2 mentioned on GitHubpytorchSyntology: not harvested report
galaxycong/hpmdubbing mentioned on GitHubpytorchSyntology: 3 samples, 3 ran (0 instrument-failed) · 0 unverified report
keonlee9420/Comprehensive-E2E-TTS mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis (arXiv:2010.05646; that paper's own run record: 17 ran (of which 11 constructed an object rather than computing a result; 14 with no instrument failure: 0 honoured, 0 violated, 14 with no contract checked; 3 where Syntology's instrument failed) · 8 unverified); nothing checks that they implement this paper's method. report
keonlee9420/Comprehensive-Transformer-TTS mentioned on GitHubpytorchSyntology: 11 samples, 8 ran (1 instrument-failed) · 3 unverified report
keonlee9420/DiffSinger mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Denoising Diffusion Probabilistic Models (arXiv:2006.11239; that paper's own run record: 186 ran (of which 81 constructed an object rather than computing a result; 142 with no instrument failure: 20 honoured, 8 violated, 114 with no contract checked; 44 where Syntology's instrument failed) · 67 unverified); nothing checks that they implement this paper's method. report
keonlee9420/Expressive-FastSpeech2 mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to FastSpeech: Fast, Robust and Controllable Text to Speech (arXiv:1905.09263; that paper's own run record: 10 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 3 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
keonlee9420/PortaSpeech mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to PortaSpeech: Portable and High-Quality Generative Text-to-Speech (arXiv:2109.15166; that paper's own run record: 12 ran (of which 1 constructed an object rather than computing a result; 9 with no instrument failure: 3 honoured, 0 violated, 6 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
komyeongjin/specdiff-gan mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to DiffGAN-TTS: High-Fidelity and Efficient Text-to-Speech with Denoising Diffusion GANs (arXiv:2201.11972; that paper's own run record: 6 ran (of which 2 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 2 violated, 1 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
majidAdibian77/ResGrad mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to ResGrad: Residual Denoising Diffusion Probabilistic Models for Text to Speech (arXiv:2212.14518; that paper's own run record: 7 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 1 honoured, 0 violated, 6 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified); nothing checks that they implement this paper's method. report
ming024/FastSpeech2 mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Attention Is All You Need (arXiv:1706.03762; that paper's own run record: 610 ran (of which 293 constructed an object rather than computing a result; 529 with no instrument failure: 45 honoured, 15 violated, 469 with no contract checked; 81 where Syntology's instrument failed) · 336 unverified); nothing checks that they implement this paper's method. report
mtresearcher/FastSpeech2 mentioned on GitHubpytorchMITSyntology: 2 samples, 2 ran (0 instrument-failed) · 0 unverified report
ndkgit339/fastspeech2-filled_pause_speech_synthesis mentioned on GitHubpytorchSyntology: 15 samples, 11 ran (1 instrument-failed) · 4 unverified report
rishikksh20/FastSpeech2 mentioned on GitHubpytorchApache-2.0Syntology: 4 samples, 4 ran (1 instrument-failed) · 0 unverified report
rishikksh20/LightSpeech mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to LightSpeech: Lightweight and Fast Text to Speech with Neural Architecture Search (arXiv:2102.04040; that paper's own run record: 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
roedoejet/fastspeech2 mentioned on GitHubpytorchMITSyntology: 3 samples, 3 ran (1 instrument-failed) · 0 unverified report
roedoejet/fastspeech2_acl2022_reproducibility mentioned on GitHubpytorchMITSyntology: not harvested report
shivammehta25/BetterFastSpeech2 mentioned on GitHubpytorchMITSyntology: 11 samples, 10 ran (4 instrument-failed) · 1 unverified report
tartunlp/transformertts mentioned on GitHubtfSyntology: 22 samples, 14 ran (2 instrument-failed) · 8 unverified report
wataru-nakata/fastspeech2-jsut mentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 1 unverified report
xcmyz/FastSpeech mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Attention Is All You Need (arXiv:1706.03762; that paper's own run record: 610 ran (of which 293 constructed an object rather than computing a result; 529 with no instrument failure: 45 honoured, 15 violated, 469 with no contract checked; 81 where Syntology's instrument failed) · 336 unverified); nothing checks that they implement this paper's method. report
yoshifumi-nakano/visual-text-to-speech mentioned on GitHubpytorchSyntology: 2 samples, 0 ran · 2 unverified report
zhangbo2008/fastSpeeck2_chinese_train mentioned on GitHubpytorchMITSyntology: not harvested report
2024-MindSpore-1/Code2 mindsporeSyntology: not harvested report
PaddlePaddle/PaddleSpeech paddleApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to A³T: Alignment-Aware Acoustic and Text Pretraining for Speech Synthesis and Editing (arXiv:2203.09690; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 5 unverified); nothing checks that they implement this paper's method. report
keonlee9420/STYLER pytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
keonlee9420/StyleSpeech pytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Meta-StyleSpeech : Multi-Speaker Adaptive Text-to-Speech Generation (arXiv:2106.03153; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 2 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
yangyucheng000/University mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution (arXiv:2409.12191; that paper's own run record: 12 ran (of which 1 constructed an object rather than computing a result; 10 with no instrument failure: 4 honoured, 1 violated, 5 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

119 samples harvested: 83 ran (of which 27 constructed an object rather than computing a result; 70 with no instrument failure: 9 honoured, 1 violated, 60 with no contract checked; 13 where Syntology's instrument failed) · 36 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

9ran · honoured contract
1ran · violated contract
12ran · our draft was wrong
1ran; the contract check could not run
60ran
36unverified

Licence: 40 of the 119 samples are pointer only, meaning Syntology does not serve that copy's text. For 7 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 16 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. For 7 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

KevinMIN95/StyleSpeech community (archive-listed) 8 samples, 6 ran (1 instrument-failed) · 2 unverified
ConvNorm models/VarianceAdaptor.py ran fingerprinted MIT (permissive) · c571f7fcdaecbe31 · report fixture: returned ConvNorm · level recorded 2026-09-22
LengthRegulator models/VarianceAdaptor.py ran MIT (permissive) · dc95367aea71c89c · report fixture: returned LengthRegulator · level recorded 2026-09-22
LinearNorm in 5 papers models/VarianceAdaptor.py ran fingerprinted MIT (permissive) · 6ebf70e53b0eb6cd · report fixture: returned LinearNorm · level recorded 2026-09-22
VarianceEmbedding models/VarianceAdaptor.py ran MIT (permissive) · c928b9499fb7fae8 · report fixture: returned VarianceEmbedding · level recorded 2026-09-22
VariancePredictor models/VarianceAdaptor.py ran MIT (permissive) · d230391d0ce5e7bb · report fixture: signature x:Tensor/3/float, mask:Tensor/2/bool · returned Tensor · level recorded 2026-09-24
get_mask_from_lengths in 7 papers models/VarianceAdaptor.py ran · our draft was wrong MIT (permissive) · 3785f1ecaa01e75b · report fixture: returned Tensor · level recorded 2026-09-22
VarianceAdaptor models/VarianceAdaptor.py unverified MIT (permissive) · 5a4bf01c36ac38a5 · report
pad in 85 papers models/VarianceAdaptor.py unverified MIT (permissive) · e77728c83dbd8597 · report
Munna-Manoj/Team6_FastSpeech2_TTS community (archive-listed) 2 samples, 0 ran · 2 unverified
LengthRegulator model/fastspeech2.py unverified MIT (permissive) · 2f458b2918186cff · report
VarianceAdaptor model/fastspeech2.py unverified MIT (permissive) · 03d514cbc289f71f · report
OlaWod/my-fastspeech2 community (archive-listed) 11 samples, 5 ran (0 instrument-failed) · 6 unverified
FFTBlock model/model.py ran · metamorphic tier: deterministic no licence file found · pointer only · 8e0eec498dd5a07b · report fixture: signature input_:Tensor/3/float, mask:Tensor/2/bool · returned tuple · level recorded 2026-09-22
MultiHeadAttention in 124 papers
census: 9 implementations from 7 papers, 9 distinct outputs across 3 buckets, one shared input per bucket
model/model.py ran · metamorphic tier: deterministic fingerprinted no licence file found · pointer only · 633b426a0ece94c3 · report fixture: signature x:Tensor/3/float, mask:NoneType/0/? · returned tuple · level recorded 2026-09-22
PositionwiseFeedForward in 36 papers model/model.py ran · metamorphic tier: invariant no licence file found · pointer only · 06c86d0f5a38e81d · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-22
PostNet model/model.py ran · metamorphic tier: invariant fingerprinted no licence file found · pointer only · 3b072d38fff02018 · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-22
ScaledDotProductAttention in 46 papers model/model.py ran no licence file found · pointer only · 0c79a4f8468a65bc · report fixture: signature q:Tensor/3/float, k:Tensor/3/float, v:Tensor/3/float, mask:Tensor/3/bool · returned tuple · level recorded 2026-09-24
Decoder in 210 papers model/model.py unverified no licence file found · pointer only · 7f7fe4818ab11be1 · report
Encoder in 278 papers model/model.py unverified no licence file found · pointer only · f1f5026c35c81551 · report
LengthRegulator model/model.py unverified no licence file found · pointer only · 279b787d9be73f04 · report
MyModel model/model.py unverified no licence file found · pointer only · 549df52c90b18370 · report
VarianceAdaptor model/model.py unverified no licence file found · pointer only · 94d2cc1c03221044 · report
VariancePredictor model/model.py unverified no licence file found · pointer only · f3ae143d1a11eba9 · report
cadia-lvl/fastspeech2 community (archive-listed) 2 samples, 1 ran (1 instrument-failed) · 1 unverified
clones in 92 papers modules.py ran · our draft was wrong MIT (permissive) · d2d810cddae9f875 · report fixture: signature module:callable/0/?, N:int/0/int · returned ModuleList · level recorded 2026-09-22
read_sentences synthesize.py unverified MIT (permissive) · 9b92f45007b02d70 · report
dathudeptrai/TensorflowTTS community (archive-listed) 21 samples, 15 ran (1 instrument-failed) · 6 unverified
TFEmbedding tensorflow_tts/models/fastspeech2.py ran Apache-2.0 (permissive) · a2368c41158a0b44 · report fixture: returned TFEmbedding · level recorded 2026-09-22
TFFastSpeechDurationPredictor tensorflow_tts/models/fastspeech2.py ran Apache-2.0 (permissive) · 115bba4b6af8210e · report fixture: signature inputs:tuple/1/EagerTensor, training:bool/0/bool · returned EagerTensor · level recorded 2026-09-24
TFFastSpeechEmbeddings tensorflow_tts/models/fastspeech2.py ran Apache-2.0 (permissive) · d03b27155c241e9b · report fixture: returned TFFastSpeechEmbeddings · level recorded 2026-09-22
TFFastSpeechIntermediate tensorflow_tts/models/fastspeech2.py ran Apache-2.0 (permissive) · 83e5495df3b935d4 · report fixture: signature inputs:tuple/1/EagerTensor · returned EagerTensor · level recorded 2026-09-24
TFFastSpeechLengthRegulator tensorflow_tts/models/fastspeech2.py ran Apache-2.0 (permissive) · 13cdbe4e5b8db636 · report fixture: returned TFFastSpeechLengthRegulator · level recorded 2026-09-22
TFFastSpeechOutput tensorflow_tts/models/fastspeech2.py ran Apache-2.0 (permissive) · 082dc93c13b04ed5 · report fixture: returned TFFastSpeechOutput · level recorded 2026-09-22
TFFastSpeechSelfAttention tensorflow_tts/models/fastspeech2.py ran Apache-2.0 (permissive) · 84c3fd85ad516f46 · report fixture: returned TFFastSpeechSelfAttention · level recorded 2026-09-22
TFFastSpeechSelfOutput tensorflow_tts/models/fastspeech2.py ran Apache-2.0 (permissive) · 6dec766f037f9f53 · report fixture: returned TFFastSpeechSelfOutput · level recorded 2026-09-22
TFFastSpeechVariantPredictor tensorflow_tts/models/fastspeech2.py ran Apache-2.0 (permissive) · 7a01e57ef50caff0 · report fixture: returned TFFastSpeechVariantPredictor · level recorded 2026-09-22
TFTacotronPostnet tensorflow_tts/models/fastspeech2.py ran Apache-2.0 (permissive) · cdf85c2097858d3f · report fixture: returned TFTacotronPostnet · level recorded 2026-09-22
gelu in 308 papers
census: 43 implementations from 74 papers, 6 distinct outputs on one shared input
tensorflow_tts/models/fastspeech2.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · 7013e9a8259233a5 · report fixture: signature x:EagerTensor/2/float · returned EagerTensor · level recorded 2026-09-22
gelu_new in 55 papers
census: 3 implementations from 7 papers, 1 distinct output on one shared input
tensorflow_tts/models/fastspeech2.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · f6c9754063b07733 · report fixture: signature x:EagerTensor/2/float · returned EagerTensor · level recorded 2026-09-22
get_initializer in 32 papers tensorflow_tts/models/fastspeech2.py ran · our draft was wrong Apache-2.0 (permissive) · dc0497b8e8f5b178 · report fixture: signature initializer_range:float/0/float · returned TruncatedNormal · level recorded 2026-09-22
mish in 95 papers
census: 5 implementations from 8 papers, 1 distinct output on one shared input
tensorflow_tts/models/fastspeech2.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · 7df891708ffe30c2 · report fixture: signature x:EagerTensor/2/float · returned EagerTensor · level recorded 2026-09-22
swish in 258 papers
census: 14 implementations from 37 papers, 2 distinct outputs on one shared input
tensorflow_tts/models/fastspeech2.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · 3f82fa4d77d98a21 · report fixture: signature x:EagerTensor/2/float · returned EagerTensor · level recorded 2026-09-22
TFFastSpeech tensorflow_tts/models/fastspeech2.py unverified Apache-2.0 (permissive) · 973dd5db4ecfb312 · report
TFFastSpeech2 tensorflow_tts/models/fastspeech2.py unverified Apache-2.0 (permissive) · d55e41791741b6a2 · report
TFFastSpeechAttention tensorflow_tts/models/fastspeech2.py unverified Apache-2.0 (permissive) · 48c60259ad977306 · report
TFFastSpeechDecoder tensorflow_tts/models/fastspeech2.py unverified Apache-2.0 (permissive) · 089f0f0fb7c5cb4d · report
TFFastSpeechEncoder tensorflow_tts/models/fastspeech2.py unverified Apache-2.0 (permissive) · 85f1a754e7f192db · report
TFFastSpeechLayer tensorflow_tts/models/fastspeech2.py unverified Apache-2.0 (permissive) · aa737e1636bc2d7b · report
galaxycong/hpmdubbing community (archive-listed) 3 samples, 3 ran (0 instrument-failed) · 0 unverified
Affective_Prosody_Adaptor model/modules.py ran · metamorphic tier: deterministic MIT (permissive) · f25cee181e5607fa · report fixture: returned Affective_Prosody_Adaptor · level recorded 2026-09-22
Conv in 30 papers
census: 8 implementations from 8 papers, 8 distinct outputs across 2 buckets, one shared input per bucket
model/modules.py ran · metamorphic tier: deterministic MIT (permissive) · d23d53aa9e38b09c · report fixture: returned Conv · level recorded 2026-09-22
VariancePredictor model/modules.py ran · metamorphic tier: deterministic MIT (permissive) · 4a7b9274be24e677 · report fixture: returned VariancePredictor · level recorded 2026-09-22
keonlee9420/Comprehensive-Transformer-TTS community (archive-listed) 11 samples, 8 ran (1 instrument-failed) · 3 unverified
BatchNorm1dTBC model/transformers/transformer_fs2.py ran · metamorphic tier: deterministic MIT (permissive) · 36b7d7479f2890e3 · report fixture: returned BatchNorm1dTBC · level recorded 2026-09-22
CustomSwish model/transformers/transformer_fs2.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · 962a3fc96762fd76 · report fixture: signature input_tensor:Tensor/2/float · returned Tensor · level recorded 2026-09-22
LayerNorm in 121 papers
census: 5 implementations from 5 papers, 5 distinct outputs across 3 buckets, one shared input per bucket
model/transformers/transformer_fs2.py ran · metamorphic tier: deterministic MIT (permissive) · dfb900695221fe2d · report fixture: returned LayerNorm · level recorded 2026-09-22
MultiheadAttention in 44 papers model/transformers/transformer_fs2.py ran · metamorphic tier: deterministic MIT (permissive) · 91289eb6932cf19a · report fixture: returned MultiheadAttention · level recorded 2026-09-22
SinusoidalPositionalEmbedding in 10 papers model/transformers/transformer_fs2.py ran MIT (permissive) · 3305668af12f291b · report fixture: signature input:Tensor/2/int, incremental_state:NoneType/0/?, timestep:NoneType/0/?, positions:NoneType/0/? · returned Tensor · level recorded 2026-09-24
Swish in 30 papers
census: 22 implementations from 19 papers, 2 distinct outputs on one shared input
model/transformers/transformer_fs2.py ran MIT (permissive) · 503cd450156c2a84 · report fixture: returned Swish · level recorded 2026-09-22
TransformerFFNLayer model/transformers/transformer_fs2.py ran · metamorphic tier: deterministic MIT (permissive) · 30e52ae5f2331859 · report fixture: returned TransformerFFNLayer · level recorded 2026-09-22
make_positions in 13 papers model/transformers/transformer_fs2.py ran · our draft was wrong fingerprinted MIT (permissive) · 3f409361d26dc11c · report fixture: signature tensor:Tensor/2/int, padding_idx:int/0/int · returned Tensor · level recorded 2026-09-22
EncSALayer model/transformers/transformer_fs2.py unverified MIT (permissive) · 68ab0d6fdb633b98 · report
FFTBlocks model/transformers/transformer_fs2.py unverified MIT (permissive) · fb0a6625708b9a0d · report
TransformerEncoderLayer in 90 papers model/transformers/transformer_fs2.py unverified MIT (permissive) · 8b3460ae89d120ba · report
mtresearcher/FastSpeech2 community (archive-listed) 2 samples, 2 ran (0 instrument-failed) · 0 unverified
get_padding in 84 papers hifigan/models.py ran · honoured contract fingerprinted MIT (permissive) here; the stored copy is pointer only · a26f85d7c72ef39a · report fixture: signature kernel_size:int/0/int, dilation:int/0/int · returned int · level recorded 2026-09-22
get_sinusoid_encoding_table in 51 papers transformer/Models.py ran · honoured contract fingerprinted MIT (permissive) here; the stored copy is pointer only · 0f8f22937463d3c5 · report fixture: signature n_position:int/0/int, d_hid:int/0/int, padding_idx:int/0/int · returned Tensor · level recorded 2026-09-22
ndkgit339/fastspeech2-filled_pause_speech_synthesis community (archive-listed) 15 samples, 11 ran (1 instrument-failed) · 4 unverified
Attention in 344 papers
census: 13 implementations from 12 papers, 10 distinct outputs across 4 buckets, one shared input per bucket
model/fastspeech2.py ran MIT (permissive) · dfedf95397e85d63 · report fixture: signature attention_hidden_state:Tensor/2/float, memory:Tensor/3/float, processed_memory:Tensor/3/float, attention_weights_cat:Tensor/3/float, mask:NoneType/0/? · returned tuple · level recorded 2026-09-24
ConvNorm model/fastspeech2.py ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · 07bf5346a131cbc4 · report fixture: returned ConvNorm · level recorded 2026-09-22
Encoder in 278 papers model/fastspeech2.py ran · metamorphic tier: deterministic MIT (permissive) · 6f4b8f2a7bb4e97e · report fixture: returned Encoder · level recorded 2026-09-22
LengthRegulator model/fastspeech2.py ran MIT (permissive) · e5a136f268af8988 · report fixture: signature x:list/1/Tensor, duration:list/1/Tensor, max_len:NoneType/0/? · returned tuple · level recorded 2026-09-24
LinearNorm in 5 papers model/fastspeech2.py ran · metamorphic tier: deterministic MIT (permissive) · ca22033a787f63fb · report fixture: returned LinearNorm · level recorded 2026-09-22
LocationLayer model/fastspeech2.py ran · metamorphic tier: deterministic MIT (permissive) · 2d670fbff60b3396 · report fixture: returned LocationLayer · level recorded 2026-09-22
PostNet model/fastspeech2.py ran · metamorphic tier: deterministic MIT (permissive) · 9b21f04966dfe516 · report fixture: returned PostNet · level recorded 2026-09-22
Prenet model/fastspeech2.py ran · metamorphic tier: deterministic MIT (permissive) · 14b683ae99e4b188 · report fixture: returned Prenet · level recorded 2026-09-22
VariancePredictor model/fastspeech2.py ran · metamorphic tier: deterministic MIT (permissive) · 5d55d1a8b50dfe0b · report fixture: returned VariancePredictor · level recorded 2026-09-22
get_mask_from_lengths in 7 papers model/fastspeech2.py ran · our draft was wrong MIT (permissive) · 02b8f4bff1530bdc · report fixture: signature lengths:Tensor/1/int, max_len:NoneType/0/? · returned Tensor · level recorded 2026-09-22
pad in 85 papers model/fastspeech2.py ran · honoured contract MIT (permissive) · 08f709a2efe5fdad · report fixture: signature input_ele:list/1/Tensor, mel_max_length:NoneType/0/? · returned Tensor · level recorded 2026-09-22
Decoder in 210 papers model/fastspeech2.py unverified MIT (permissive) · feecea3092482d06 · report
FastSpeech2 model/fastspeech2.py unverified MIT (permissive) · 53bba144fff59272 · report
JDIT model/fastspeech2.py unverified MIT (permissive) · 39322f2fda5b6bd4 · report
VarianceAdaptor model/fastspeech2.py unverified MIT (permissive) · b3c4279a4be411d9 · report
rishikksh20/FastSpeech2 community (archive-listed) 4 samples, 4 ran (1 instrument-failed) · 0 unverified
evaluate in 514 papers evaluation.py ran Apache-2.0 (permissive) · 644e4bb8a34cbee8 · report fixture: signature hp:DummyConfig/0/?, validloader:list/1/tuple, model:callable/0/? · returned tuple · level recorded 2026-09-24
get_std_opt in 6 papers core/optimizer.py ran Apache-2.0 (permissive) · 09d756d182147053 · report fixture: signature model:callable/0/?, d_model:int/0/int, warmup:int/0/int, factor:float/0/float · returned NoamOpt · level recorded 2026-09-24
process_paragraph inference.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 141ed5512c50d7db · report fixture: signature para:str/0/str · returned list · level recorded 2026-09-22
repeat in 22 papers core/modules.py ran Apache-2.0 (permissive) · c6691c20de5ab4d1 · report fixture: signature N:int/0/int, fn:callable/0/? · returned MultiSequential · level recorded 2026-09-24
roedoejet/fastspeech2 community (archive-listed) 3 samples, 3 ran (1 instrument-failed) · 0 unverified
get_punctuation_features features.py ran fingerprinted MIT (permissive) · d5ee014ab3ada236 · report fixture: signature text:str/0/str · returned list · level recorded 2026-09-24
get_tone_features features.py ran MIT (permissive) · 61ccd4fca5223693 · report fixture: returned list · level recorded 2026-09-26
read_lexicon in 16 papers synthesize.py ran · our draft was wrong MIT (permissive) · 2b5c06c1ba593c3c · report fixture: signature lex_path:str/0/str · returned dict · level recorded, date not recorded
shivammehta25/BetterFastSpeech2 community (archive-listed) 11 samples, 10 ran (4 instrument-failed) · 1 unverified
convert_pad_shape in 15 papers fs2/utils/model.py ran MIT (permissive) · ecfa879e1ca09709 · report fixture: signature pad_shape:list/1/tuple · returned list · level recorded 2026-09-24
discriminator_loss in 38 papers fs2/hifigan/models.py ran MIT (permissive) here; the stored copy is pointer only · 7137577cbef51217 · report fixture: signature disc_real_outputs:list/1/Tensor, disc_generated_outputs:list/1/Tensor · returned tuple · level recorded 2026-09-24
dynamic_range_compression in 31 papers fs2/hifigan/meldataset.py ran fingerprinted MIT (permissive) · be031b15a988a556 · report fixture: signature x:ndarray/1/float, C:int/0/int, clip_val:float/0/float · returned ndarray · level recorded 2026-09-25
dynamic_range_decompression in 15 papers fs2/hifigan/meldataset.py ran fingerprinted MIT (permissive) · 9e30e7d510c9d341 · report fixture: signature x:ndarray/1/float, C:float/0/float · returned ndarray · level recorded 2026-09-25
feature_loss in 27 papers fs2/hifigan/models.py ran MIT (permissive) here; the stored copy is pointer only · e453b51f0ed5fb28 · report fixture: signature fmap_r:list/1/list, fmap_g:list/1/list · returned Tensor · level recorded 2026-09-24
generator_loss in 32 papers fs2/hifigan/models.py ran · our draft was wrong MIT (permissive) here; the stored copy is pointer only · 1a9d74439d969cfd · report fixture: signature disc_outputs:list/1/Tensor · returned tuple · level recorded 2026-09-24
load_checkpoint in 277 papers fs2/hifigan/xutils.py ran MIT (permissive) · acdbb0b4f93fe3c5 · report fixture: signature filepath:str/0/str, device:str/0/str · returned dict · level recorded 2026-09-24
load_wav in 14 papers fs2/hifigan/meldataset.py ran · our draft was wrong MIT (permissive) here; the stored copy is pointer only · 8043eaefe4c09c1f · report fixture: signature full_path:str/0/str · returned tuple · level recorded 2026-09-24
plot_spectrogram in 10 papers fs2/hifigan/xutils.py ran · our draft was wrong MIT (permissive) · 8520e6e61b8bf789 · report fixture: signature spectrogram:ndarray/2/float · returned Figure · level recorded 2026-09-25
sequence_mask in 55 papers
census: 12 implementations from 15 papers, 5 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 1 of 2, arg 1: rank 1, kind int, dtype int64
fs2/utils/model.py ran · our draft was wrong MIT (permissive) here; the stored copy is pointer only · 88bf809d1bc9db4c · report fixture: signature length:Tensor/1/int, max_length:NoneType/0/? · returned Tensor · level recorded 2026-09-22
generate_path fs2/utils/model.py unverified MIT (permissive) · a37351e34576962b · report
tartunlp/transformertts community (archive-listed) 22 samples, 14 ran (2 instrument-failed) · 8 unverified
CNNDropout src/transformer_tts/model/forward_model.py ran fingerprinted licence not identified · pointer only · d5e9a425417e7d5f · report fixture: signature inputs:EagerTensor/3/float, training:bool/0/bool · returned EagerTensor · level recorded 2026-09-24
CNNResNorm src/transformer_tts/model/forward_model.py ran licence not identified · pointer only · 8882509c8098093d · report fixture: signature inputs:EagerTensor/3/float, training:bool/0/bool · returned EagerTensor · level recorded 2026-09-24
MultiHeadAttention in 124 papers
census: 9 implementations from 7 papers, 9 distinct outputs across 3 buckets, one shared input per bucket
src/transformer_tts/model/forward_model.py ran · metamorphic tier: deterministic licence not identified · pointer only · 0d6a9c2615231e6a · report fixture: signature v:EagerTensor/3/float, k:EagerTensor/3/float, q_in:EagerTensor/3/float, mask:NoneType/0/?, training:bool/0/bool · returned tuple · level recorded 2026-09-22
Phonemizer src/transformer_tts/model/forward_model.py ran · metamorphic tier: deterministic licence not identified · pointer only · 3bbdc525d00b1109 · report fixture: signature text:list/1/str · returned list · level recorded 2026-09-22
ScaledDotProductAttention in 46 papers src/transformer_tts/model/forward_model.py ran · metamorphic tier: deterministic licence not identified · pointer only · 27a396b3518d14df · report fixture: signature inputs:tuple/1/EagerTensor · returned tuple · level recorded 2026-09-22
SelfAttentionResNorm src/transformer_tts/model/forward_model.py ran fingerprinted licence not identified · pointer only · ac9cdd8c4dc1592a · report fixture: signature x:EagerTensor/3/float, training:bool/0/bool, mask:NoneType/0/? · returned tuple · level recorded 2026-09-24
StatPredictor src/transformer_tts/model/forward_model.py ran · metamorphic tier: invariant licence not identified · pointer only · fc2705c21133a77b · report fixture: signature x:EagerTensor/3/float, training:bool/0/bool, mask:EagerTensor/3/float · returned EagerTensor · level recorded 2026-09-22
TextToTokens src/transformer_tts/model/forward_model.py ran licence not identified · pointer only · 21b9024092bb96f5 · report fixture: signature input_text:list/1/str, speaker_id:int/0/int · returned EagerTensor · level recorded 2026-09-24
Tokenizer in 16 papers src/transformer_tts/model/forward_model.py ran · metamorphic tier: well formed licence not identified · pointer only · 6e41883c2988e191 · report fixture: signature sentence:list/1/str, speaker_id:list/1/int · returned list · level recorded 2026-09-22
create_encoder_padding_mask src/transformer_tts/model/forward_model.py ran · honoured contract fingerprinted licence not identified · pointer only · db55b6a434821c7d · report fixture: signature seq:EagerTensor/2/int · returned EagerTensor · level recorded 2026-09-22
create_mel_padding_mask src/transformer_tts/model/forward_model.py ran; the contract check could not run (RAISES) fingerprinted licence not identified · pointer only · 9975ceb350d936d5 · report fixture: signature seq:EagerTensor/3/float · returned EagerTensor · level recorded 2026-09-22
masked_mean_absolute_error src/transformer_tts/model/forward_model.py ran · violated contract licence not identified · pointer only · e9d21f4787357859 · report fixture: signature targets:EagerTensor/2/float, logits:EagerTensor/2/float, mask_value:int/0/int, mask:NoneType/0/? · returned EagerTensor · level recorded 2026-09-24
positional_encoding in 63 papers
census: 5 implementations from 6 papers, 4 distinct outputs on one shared input
src/transformer_tts/model/forward_model.py ran · honoured contract licence not identified · pointer only · 4c331ef700eda3c3 · report fixture: signature position:int/0/int, model_dim:int/0/int · returned EagerTensor · level recorded 2026-09-22
weighted_sum_losses src/transformer_tts/model/forward_model.py ran · our draft was wrong licence not identified · pointer only · e5864b560f323650 · report fixture: signature targets:list/1/EagerTensor, pred:list/1/EagerTensor, loss_functions:list/1/callable, coeffs:list/1/float · returned tuple · level recorded 2026-09-24
Expand src/transformer_tts/model/forward_model.py unverified licence not identified · pointer only · 3f2a99ca2b26973e · report
FFNResNorm src/transformer_tts/model/forward_model.py unverified licence not identified · pointer only · 10a6bdb9406b5915 · report
ForwardTransformer src/transformer_tts/model/forward_model.py unverified licence not identified · pointer only · c992ed713570824f · report
Multispeaker src/transformer_tts/model/forward_model.py unverified licence not identified · pointer only · 2f5af7997249b4d8 · report
SelfAttentionBlocks src/transformer_tts/model/forward_model.py unverified licence not identified · pointer only · 56f2decdf988a415 · report
SelfAttentionConvBlock src/transformer_tts/model/forward_model.py unverified licence not identified · pointer only · f672e5379057df58 · report
SelfAttentionDenseBlock src/transformer_tts/model/forward_model.py unverified licence not identified · pointer only · 5fbb588b59a8750e · report
TransposedCNNResNorm src/transformer_tts/model/forward_model.py unverified licence not identified · pointer only · 66a2f904c23b90c1 · report
wataru-nakata/fastspeech2-jsut community (archive-listed) 1 sample, 0 ran · 1 unverified
FastSpeech2 model/fastspeech2.py unverified MIT (permissive) · f5b9382704417d86 · report
yoshifumi-nakano/visual-text-to-speech community (archive-listed) 2 samples, 0 ran · 2 unverified
VarianceAdaptor model/vtts.py unverified MIT (permissive) · 6c171089db79b527 · report
vTTS model/vtts.py unverified MIT (permissive) · 482f6777791c0248 · report
keonlee9420/STYLER community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
LengthRegulator modules.py ran MIT (permissive) · b0463abb38c1cc46 · report fixture: signature x:list/1/Tensor, duration:list/1/Tensor, max_len:NoneType/0/? · returned tuple · level recorded 2026-09-24

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

11 assemblies on this paper: 11 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

11 sibling assemblies for FastSpeech 2 (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for FastSpeech 2 (graph method: FastSpeech 2). 12 bricks; individually on a fixture: 9 ran (of which 6 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 2 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

one of 11 siblings

  1. FFTBlocks entry · level 0 · unverified · fb0a6625708b9a0d
  2. BatchNorm1dTBC model · level 1 · ran · 36b7d7479f2890e3
  3. SinusoidalPositionalEmbedding helper · level 1 · ran · 3305668af12f291b
  4. TransformerEncoderLayer model · level 0 · unverified · 8b3460ae89d120ba
  5. make_positions helper · level 1 · ran · our draft was wrong · 3f409361d26dc11c
  6. EncSALayer helper · level 0 · unverified · 68ab0d6fdb633b98
  7. LayerNorm helper · level 1 · ran · dfb900695221fe2d
  8. MultiheadAttention model · level 1 · ran · 91289eb6932cf19a
  9. TransformerFFNLayer model · level 1 · ran · 30e52ae5f2331859
  10. CustomSwish helper · level 1 · ran · 962a3fc96762fd76
  11. Linear helper · level 1 · ran · our draft was wrong · 8cd8cb0d1e9e63e4
  12. Swish helper · level 1 · ran · 503cd450156c2a84

harvested Assembly for FastSpeech 2 (graph method: FastSpeech 2). 3 bricks; individually on a fixture: 3 ran (of which 3 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; every one of the 3 samples that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 11 siblings

  1. Affective_Prosody_Adaptor entry · level 1 · ran · f25cee181e5607fa
  2. VariancePredictor model · level 1 · ran · 4a7b9274be24e677
  3. Conv model · level 1 · ran · d23d53aa9e38b09c

harvested Assembly for FastSpeech 2 (graph method: FastSpeech 2). 16 bricks; individually on a fixture: 12 ran (of which 8 constructed an object rather than computing a result; 11 with no instrument failure: 1 honoured, 0 violated, 10 with no contract checked; 1 where Syntology's instrument failed) · 4 unverified. The assembly has not been run end-to-end.

one of 11 siblings

  1. FastSpeech2 entry · level 0 · unverified · 53bba144fff59272
  2. Decoder helper · level 0 · unverified · feecea3092482d06
  3. Encoder helper · level 1 · ran · 6f4b8f2a7bb4e97e
  4. JDIT helper · level 0 · unverified · 39322f2fda5b6bd4
  5. PostNet helper · level 1 · ran · 9b21f04966dfe516
  6. VarianceAdaptor model · level 0 · unverified · b3c4279a4be411d9
  7. get_mask_from_lengths helper · level 1 · ran · our draft was wrong · 02b8f4bff1530bdc
  8. Attention helper · level 1 · ran · dfedf95397e85d63
  9. LinearNorm helper · level 1 · ran · ca22033a787f63fb
  10. Prenet helper · level 1 · ran · 14b683ae99e4b188
  11. ConvNorm helper · level 1 · ran · 07bf5346a131cbc4
  12. LengthRegulator model · level 1 · ran · e5a136f268af8988
  13. VariancePredictor model · level 1 · ran · 5d55d1a8b50dfe0b
  14. LocationLayer helper · level 1 · ran · 2d670fbff60b3396
  15. pad helper · level 1 · ran · honoured contract · 08f709a2efe5fdad
  16. Conv helper · level 1 · ran · d23d53aa9e38b09c

harvested Assembly for FastSpeech 2 (graph method: FastSpeech 2). 2 bricks; individually on a fixture: 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 11 siblings

  1. LengthRegulator entry · level 1 · ran · b0463abb38c1cc46
  2. pad helper · level 1 · ran · honoured contract · 08f709a2efe5fdad

harvested Assembly for FastSpeech 2 (graph method: FastSpeech 2). 21 bricks; individually on a fixture: 15 ran (of which 8 constructed an object rather than computing a result; 14 with no instrument failure: 4 honoured, 0 violated, 10 with no contract checked; 1 where Syntology's instrument failed) · 6 unverified. The assembly has not been run end-to-end.

one of 11 siblings

  1. TFFastSpeech2 entry · level 0 · unverified · d55e41791741b6a2
  2. TFFastSpeech helper · level 0 · unverified · 973dd5db4ecfb312
  3. TFFastSpeechVariantPredictor model · level 1 · ran · 7a01e57ef50caff0
  4. TFFastSpeechDecoder model · level 0 · unverified · 089f0f0fb7c5cb4d
  5. TFFastSpeechDurationPredictor model · level 1 · ran · 115bba4b6af8210e
  6. TFFastSpeechEmbeddings model · level 1 · ran · d03b27155c241e9b
  7. TFFastSpeechEncoder model · level 0 · unverified · 85f1a754e7f192db
  8. TFFastSpeechLengthRegulator model · level 1 · ran · 13cdbe4e5b8db636
  9. TFTacotronPostnet helper · level 1 · ran · cdf85c2097858d3f
  10. get_initializer helper · level 1 · ran · our draft was wrong · dc0497b8e8f5b178
  11. TFEmbedding helper · level 1 · ran · a2368c41158a0b44
  12. TFFastSpeechLayer helper · level 0 · unverified · aa737e1636bc2d7b
  13. TFFastSpeechAttention helper · level 0 · unverified · 48c60259ad977306
  14. TFFastSpeechIntermediate helper · level 1 · ran · 83e5495df3b935d4
  15. TFFastSpeechOutput helper · level 1 · ran · 082dc93c13b04ed5
  16. TFFastSpeechSelfAttention helper · level 1 · ran · 84c3fd85ad516f46
  17. TFFastSpeechSelfOutput helper · level 1 · ran · 6dec766f037f9f53
  18. gelu helper · level 1 · ran · honoured contract · 7013e9a8259233a5
  19. gelu_new helper · level 1 · ran · honoured contract · f6c9754063b07733
  20. mish helper · level 1 · ran · honoured contract · 7df891708ffe30c2
  21. swish helper · level 1 · ran · honoured contract · 3f82fa4d77d98a21

harvested Assembly for FastSpeech 2 (graph method: FastSpeech 2). 9 bricks; individually on a fixture: 7 ran (of which 4 constructed an object rather than computing a result; 6 with no instrument failure: 1 honoured, 0 violated, 5 with no contract checked; 1 where Syntology's instrument failed) · 2 unverified. The assembly has not been run end-to-end.

one of 11 siblings

  1. VarianceAdaptor entry · level 0 · unverified · 5a4bf01c36ac38a5
  2. LengthRegulator model · level 1 · ran · dc95367aea71c89c
  3. VarianceEmbedding model · level 1 · ran · c928b9499fb7fae8
  4. VariancePredictor model · level 1 · ran · d230391d0ce5e7bb
  5. get_mask_from_lengths helper · level 1 · ran · our draft was wrong · 3785f1ecaa01e75b
  6. get_sinusoid_encoding_table helper · level 1 · ran · honoured contract · 0f8f22937463d3c5
  7. pad helper · level 0 · unverified · e77728c83dbd8597
  8. ConvNorm helper · level 1 · ran · c571f7fcdaecbe31
  9. LinearNorm helper · level 1 · ran · 6ebf70e53b0eb6cd

harvested Assembly for FastSpeech 2 (graph method: FastSpeech 2). 16 bricks; individually on a fixture: 12 ran (of which 8 constructed an object rather than computing a result; 11 with no instrument failure: 1 honoured, 0 violated, 10 with no contract checked; 1 where Syntology's instrument failed) · 4 unverified. The assembly has not been run end-to-end.

one of 11 siblings

  1. vTTS entry · level 0 · unverified · 482f6777791c0248
  2. Decoder model · level 0 · unverified · feecea3092482d06
  3. Encoder model · level 1 · ran · 6f4b8f2a7bb4e97e
  4. JDIT helper · level 0 · unverified · 39322f2fda5b6bd4
  5. PostNet helper · level 1 · ran · 9b21f04966dfe516
  6. VarianceAdaptor model · level 0 · unverified · 6c171089db79b527
  7. get_mask_from_lengths helper · level 1 · ran · our draft was wrong · 02b8f4bff1530bdc
  8. Attention helper · level 1 · ran · dfedf95397e85d63
  9. LinearNorm helper · level 1 · ran · ca22033a787f63fb
  10. Prenet helper · level 1 · ran · 14b683ae99e4b188
  11. ConvNorm helper · level 1 · ran · 07bf5346a131cbc4
  12. LengthRegulator model · level 1 · ran · e5a136f268af8988
  13. VariancePredictor model · level 1 · ran · 5d55d1a8b50dfe0b
  14. LocationLayer helper · level 1 · ran · 2d670fbff60b3396
  15. pad helper · level 1 · ran · honoured contract · 08f709a2efe5fdad
  16. Conv helper · level 1 · ran · d23d53aa9e38b09c

harvested Assembly for FastSpeech 2 (graph method: FastSpeech 2). 16 bricks; individually on a fixture: 12 ran (of which 8 constructed an object rather than computing a result; 11 with no instrument failure: 1 honoured, 0 violated, 10 with no contract checked; 1 where Syntology's instrument failed) · 4 unverified. The assembly has not been run end-to-end.

one of 11 siblings

  1. FastSpeech2 entry · level 0 · unverified · f5b9382704417d86
  2. Decoder model · level 0 · unverified · feecea3092482d06
  3. Encoder model · level 1 · ran · 6f4b8f2a7bb4e97e
  4. JDIT helper · level 0 · unverified · 39322f2fda5b6bd4
  5. PostNet helper · level 1 · ran · 9b21f04966dfe516
  6. VarianceAdaptor model · level 0 · unverified · b3c4279a4be411d9
  7. get_mask_from_lengths helper · level 1 · ran · our draft was wrong · 02b8f4bff1530bdc
  8. Attention helper · level 1 · ran · dfedf95397e85d63
  9. LinearNorm helper · level 1 · ran · ca22033a787f63fb
  10. Prenet helper · level 1 · ran · 14b683ae99e4b188
  11. ConvNorm helper · level 1 · ran · 07bf5346a131cbc4
  12. LengthRegulator helper · level 1 · ran · e5a136f268af8988
  13. VariancePredictor helper · level 1 · ran · 5d55d1a8b50dfe0b
  14. LocationLayer helper · level 1 · ran · 2d670fbff60b3396
  15. pad helper · level 1 · ran · honoured contract · 08f709a2efe5fdad
  16. Conv helper · level 1 · ran · d23d53aa9e38b09c

harvested Assembly for FastSpeech 2 (graph method: FastSpeech 2). 23 bricks; individually on a fixture: 15 ran (of which 0 constructed an object rather than computing a result; 13 with no instrument failure: 2 honoured, 2 violated, 9 with no contract checked; 2 where Syntology's instrument failed) · 8 unverified. The assembly has not been run end-to-end.

one of 11 siblings

  1. ForwardTransformer entry · level 0 · unverified · c992ed713570824f
  2. Expand helper · level 0 · unverified · 3f2a99ca2b26973e
  3. Multispeaker helper · level 0 · unverified · 2f5af7997249b4d8
  4. SelfAttentionBlocks model · level 0 · unverified · 56f2decdf988a415
  5. StatPredictor model · level 1 · ran · fc2705c21133a77b
  6. TextToTokens helper · level 1 · ran · 21b9024092bb96f5
  7. create_encoder_padding_mask helper · level 1 · ran · honoured contract · db55b6a434821c7d
  8. create_mel_padding_mask helper · level 1 · ran; the contract check could not run (RAISES) · 9975ceb350d936d5
  9. masked_mean_absolute_error helper · level 1 · ran · violated contract · e9d21f4787357859
  10. weighted_sum_losses helper · level 1 · ran · our draft was wrong · e5864b560f323650
  11. SelfAttentionConvBlock model · level 0 · unverified · f672e5379057df58
  12. SelfAttentionDenseBlock model · level 0 · unverified · 5fbb588b59a8750e
  13. positional_encoding helper · level 1 · ran · honoured contract · 4c331ef700eda3c3
  14. CNNDropout helper · level 1 · ran · d5e9a425417e7d5f
  15. Phonemizer helper · level 1 · ran · 3bbdc525d00b1109
  16. Tokenizer helper · level 1 · ran · 6e41883c2988e191
  17. CNNResNorm model · level 1 · ran · 8882509c8098093d
  18. SelfAttentionResNorm model · level 1 · ran · ac9cdd8c4dc1592a
  19. TransposedCNNResNorm helper · level 0 · unverified · 66a2f904c23b90c1
  20. FFNResNorm model · level 0 · unverified · 10a6bdb9406b5915
  21. get_angles helper · level 1 · ran · violated contract · 2cf1092de7ac3279
  22. MultiHeadAttention model · level 1 · ran · 0d6a9c2615231e6a
  23. ScaledDotProductAttention model · level 1 · ran · 27a396b3518d14df

harvested Assembly for FastSpeech 2 (graph method: FastSpeech 2). 11 bricks; individually on a fixture: 5 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 0 where Syntology's instrument failed) · 6 unverified. The assembly has not been run end-to-end.

one of 11 siblings

  1. MyModel entry · level 0 · unverified · 549df52c90b18370
  2. Decoder model · level 0 · unverified · 7f7fe4818ab11be1
  3. Encoder model · level 0 · unverified · f1f5026c35c81551
  4. PostNet helper · level 1 · ran · 3b072d38fff02018
  5. VarianceAdaptor model · level 0 · unverified · 94d2cc1c03221044
  6. FFTBlock helper · level 1 · ran · 8e0eec498dd5a07b
  7. LengthRegulator model · level 0 · unverified · 279b787d9be73f04
  8. VariancePredictor model · level 0 · unverified · f3ae143d1a11eba9
  9. MultiHeadAttention helper · level 1 · ran · 633b426a0ece94c3
  10. PositionwiseFeedForward helper · level 1 · ran · 06c86d0f5a38e81d
  11. ScaledDotProductAttention helper · level 1 · ran · 0c79a4f8468a65bc

harvested Assembly for FastSpeech 2 (graph method: FastSpeech 2). 4 bricks; individually on a fixture: 2 ran (of which 2 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified; every one of the 2 samples that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 11 siblings

  1. VarianceAdaptor entry · level 0 · unverified · 03d514cbc289f71f
  2. LengthRegulator model · level 0 · unverified · 2f458b2918186cff
  3. VariancePredictor model · level 1 · ran · 4a7b9274be24e677
  4. Conv helper · level 1 · ran · d23d53aa9e38b09c

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2006.04558") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Knowledge DistillationSpeech SynthesisText to SpeechText-To-Speech Synthesistext-to-speech

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Text-To-Speech Synthesis LJSpeech FastSpeech 2 + HiFiGAN Audio Quality MOS 4.32 #6 of 16 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: FastSpeech 2, FastSpeech 2s

1x1 ConvolutionAttentionBatch NormalizationConvolutionDense ConnectionsDilated ConvolutionDropoutFastSpeech 2FastSpeech 2sGAN Least Squares LossKnowledge DistillationLSGANLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerReLUResidual ConnectionSoftmax

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