Codesota · Speech · Text-to-Speech · LJ SpeechTasks/Speech/Text-to-Speech
Text-to-Speech · benchmark dataset · 2017 · EN

The LJ Speech Dataset.

13,100 short audio clips of a single speaker reading passages from non-fiction books. Standard benchmark for single-speaker TTS.

Paper ↗Download datasetSubmit a result ↵
§ 01 · Leaderboard

Best published scores.

5 results indexed across 1 metric. Shaded row marks current SOTA; ties broken by submission date.

Primary
mos · higher is better
mos· primary
5 rows
#ModelOrgSubmittedPaper / codemos
01VALL-E 2MicrosoftJun 2024VALL-E 2: Neural Codec Language Models are Human Parity …4.61
02StyleTTS 2OSSColumbia UniversityJun 2023StyleTTS 2: Towards Human-Level Text-to-Speech through S…4.55
03VITSOSSKakaoJun 2021VITS: Conditional Variational Autoencoder with Adversari…4.43
04VoiceboxMeta AIJun 2023Voicebox: Text-Guided Multilingual Universal Speech Gene…4.30
05XTTS v2OSSCoqui AIApr 2023XTTS: A Massively Multilingual Zero-Shot Text-to-Speech …4.21
Fig 2 · Rows sorted by score within each metric. Shaded row marks SOTA. Dates reflect model or paper release where available, otherwise the date Codesota accessed the source.
§ 03 · Progress

3 steps
of state of the art.

Each row below marks a model that broke the previous record on mos. Intermediate submissions are kept in the leaderboard above; only SOTA-setting entries are re-listed here.

Higher scores win. Each subsequent entry improved upon the previous best.

SOTA line · mos
  1. Jun 11, 2021VITSKakao4.43
  2. Jun 12, 2023StyleTTS 2Columbia University4.55
  3. Jun 8, 2024VALL-E 2Microsoft4.61
Fig 3 · SOTA-setting models only. 3 entries span Jun 2021 → Jun 2024.
§ 04 · Literature

5 papers
tied to this benchmark.

Every paper below corresponds to at least one row in the leaderboard above. Click through for the arXiv preprint and, when available, the reference implementation.

§ 06 · Contribute

Have a score that beats
this table?

Submit a checkpoint and a reproduction script. We will run it, publish the score, and — if it takes the top — annotate the step on the progress chart with your name.

Submit a result ↵Read submission guide
What a submission needs
  • 01A public checkpoint or API endpoint
  • 02A reproduction script with frozen commit + seed
  • 03Declared evaluation environment (Python, deps)
  • 04One row per metric declared by this dataset
  • 05A contact so we can follow up on discrepancies
LJ Speech — Text-to-Speech benchmark · Codesota | CodeSOTA