# Text-to-Speech on ParaNet 100 sentences (test)

Source: [Text-to-Speech on ParaNet 100 sentences (test)](<https://www.sota2.com/research/sota/text-to-speech-on-paranet-100-sentences-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T06:02:06.708Z

Evaluation of Text-to-Speech synthesis quality using the test split of the ParaNet 100 sentences dataset, focusing on specific error analysis metrics such as repeats, mispronunciations, and skips.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Text-to-Speech |
| Dataset (including any stated split) | ParaNet 100 sentences (test) |
| Metric | Repeat Errors |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 6 of 6 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Repeat Errors](<https://www.sota2.com/research/sota/text-to-speech-on-paranet-100-sentences-test.md?metric=Repeat%20Errors>): Lower is better; unit: SCALAR.

- [Mispronunciation Errors](<https://www.sota2.com/research/sota/text-to-speech-on-paranet-100-sentences-test.md?metric=Mispronunciation%20Errors>): Lower is better; unit: SCALAR.

- [Skipped Content Count](<https://www.sota2.com/research/sota/text-to-speech-on-paranet-100-sentences-test.md?metric=Skipped%20Content%20Count>): Lower is better; unit: SCALAR.

- [Total Errors](<https://www.sota2.com/research/sota/text-to-speech-on-paranet-100-sentences-test.md?metric=Total%20Errors>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Repeat Errors | Result date | Report paper |
| --- | --- | --- | --- |
| Tacotron 2 | 0 | Not recorded | 68fa68a01e209131cea1d35e |
| Glow-TTS | 0 | Not recorded | 68fa68a01e209131cea1d35e |
| DeepVoice 3 | 1 | Not recorded | 68fa68a01e209131cea1d35e |
| ParaNet | 1 | Not recorded | 68fa68a01e209131cea1d35e |
| ParaNet | 2 | Not recorded | 68fa68a01e209131cea1d35e |
| DeepVoice 3 | 12 | Not recorded | 68fa68a01e209131cea1d35e |

### Tacotron 2

Repeat Errors: 0. Result ID: 698d79d88b517eb43321c79d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa68a01e209131cea1d35e.md>)

Recorded evaluation configuration: {"Attention Mask":false}. This record does not assert independent reproduction.

### Glow-TTS

Repeat Errors: 0. Result ID: 698d79d88b517eb43321c79e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa68a01e209131cea1d35e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68a01e209131cea1d35e.md>)

[Implementation](<https://github.com/jaywalnut310/glow-tts>)

Recorded evaluation configuration: {"Sampling Temperature (T)":0.333,"Attention Mask":false}. This record does not assert independent reproduction.

### DeepVoice 3

Repeat Errors: 1. Result ID: 698d79d88b517eb43321c79a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa68a01e209131cea1d35e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67d51e209131cea18cc1.md>)

Recorded evaluation configuration: {"Attention Mask":true}. This record does not assert independent reproduction.

### ParaNet

Repeat Errors: 1. Result ID: 698d79d88b517eb43321c79b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa68a01e209131cea1d35e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67d51e209131cea18cc1.md>)

Recorded evaluation configuration: {"Attention Mask":false}. This record does not assert independent reproduction.

### ParaNet

Repeat Errors: 2. Result ID: 698d79d88b517eb43321c79c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa68a01e209131cea1d35e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67d51e209131cea18cc1.md>)

Recorded evaluation configuration: {"Attention Mask":true}. This record does not assert independent reproduction.

### DeepVoice 3

Repeat Errors: 12. Result ID: 698d79d88b517eb43321c799.

[Reporting paper](<https://www.sota2.com/research/paper/68fa68a01e209131cea1d35e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67d51e209131cea18cc1.md>)

Recorded evaluation configuration: {"Attention Mask":false}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.