SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2203.16852 · 2022

JETS: Jointly Training FastSpeech2 and HiFi-GAN for End to End Text to Speech

arXiv · PDF · Open in the Atlas

Code that ran

2 of 2 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 0 failures of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2203.16852.json · badge: /api/ran/2203.16852.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
imdanboy/jets canonical (classifier label)
classifier basis not recorded on the link
2 ran · 0 no recorded run
FunctionStatusWhere it lives
average_by_duration Ran
ran on a synthesized input
imdanboy/jets/espnet2/gan_tts/jets/alignments.py
code served (permissive licence) · get_code("a722c89f6a70796a")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family int, shape (2, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Create small test tensors
# ds: (B, T_text) - token durations
# xs: (B, T_feats) - feature sequences
# text_lengths: (B,) - actual text lengths per batch
# feats_lengths: (B,) - actual feature lengths per batch

B = 2  # batch size
T_text = 3  # text length
T_feats = 7  # feature length
feat_dim = 4  # feature dimension

# Token durations - must be positive integers that sum to feature length
ds = torch.tensor([[2, 3, 2], [2, 2, 3]], dtype=torch.int64)

# Feature sequences - shape (B, T_feats, feat_dim) or (B, T_feats)
# Based on the function, xs needs to support slicing and .mean()
xs = torch.randn(B, T_feats, feat_dim, dtype=torch.float32)

# Text lengths - actual lengths for each batch item
text_lengths = torch.tensor([3, 3], dtype=torch.int64)

# Feature lengths - actual feature lengths for each batch item
feats_lengths = torch.tensor([7, 7], dtype=torch.int64)

ARGS = {
    "ds": ds,
    "xs": xs,
    "text_lengths": text_lengths,
    "feats_lengths": feats_lengths,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numba, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family int, shape (2, 3) — shape (2, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
viterbi_decode Ran
ran on a synthesized input
imdanboy/jets/espnet2/gan_tts/jets/alignments.py
code served (permissive licence) · get_code("5299eba669841b1d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Create small batch of log probability attention matrices
# Shape: (B, T_feats, T_text)
# B=2 (batch size), T_feats=5 (feature time steps), T_text=4 (text time steps)
log_p_attn = torch.randn((2, 5, 4), dtype=torch.float32)

# Text lengths for each batch item
text_lengths = torch.tensor([4, 3], dtype=torch.long)

# Feature lengths for each batch item
feats_lengths = torch.tensor([5, 4], dtype=torch.long)

ARGS = {
    "log_p_attn": log_p_attn,
    "text_lengths": text_lengths,
    "feats_lengths": feats_lengths,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numba, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 2 samples lifted from this paper's linked repositories, 2 carry a function identity and 0 do not. Of the 2 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 2 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

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

Abstract

In neural text-to-speech (TTS), two-stage system or a cascade of separately learned models have shown synthesis quality close to human speech. For example, FastSpeech2 transforms an input text to a mel-spectrogram and then HiFi-GAN generates a raw waveform from a mel-spectogram where they are called an acoustic feature generator and a neural vocoder respectively. However, their training pipeline is somewhat cumbersome in that it requires a fine-tuning and an accurate speech-text alignment for optimal performance. In this work, we present end-to-end text-to-speech (E2E-TTS) model which has a simplified training pipeline and outperforms a cascade of separately learned models. Specifically, our proposed model is jointly trained FastSpeech2 and HiFi-GAN with an alignment module. Since there is no acoustic feature mismatch between training and inference, it does not requires fine-tuning. Furthermore, we remove dependency on an external speech-text alignment tool by adopting an alignment learning objective in our joint training framework. Experiments on LJSpeech corpus shows that the proposed model outperforms publicly available, state-of-the-art implementations of ESPNet2-TTS on subjective evaluation (MOS) and some objective evaluations.

For agents

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

get_harvested_code_for_paper("2203.16852")
get_code_for_paper("2203.16852")
have("2203.16852")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2203.16852.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2203.16852.svg)](https://syntology.ai/paper/2203.16852)

Connect an agent — have() is free.