# Text-to-Speech on North American English TTS dataset single-speaker (test)

Source: [Text-to-Speech on North American English TTS dataset single-speaker (test)](<https://www.sota2.com/research/sota/text-to-speech-on-north-american-english-tts-dataset-single-speaker-test>)

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

Data updated: 2026-02-26T05:50:04.303Z

Evaluation of Text-to-Speech synthesis quality using the single-speaker test split of the North American English TTS dataset, typically measured using Mean Opinion Score (MOS).

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) | North American English TTS dataset single-speaker (test) |
| Metric | MOS |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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

- [MOS](<https://www.sota2.com/research/sota/text-to-speech-on-north-american-english-tts-dataset-single-speaker-test.md?metric=MOS>): Higher is better; unit: SCALAR.

## Recorded results

| Method | MOS | Result date | Report paper |
| --- | --- | --- | --- |
| Natural | 4.55 | Not recorded | 68fa62bf1e209131ce9fb1ba |
| Natural | 4.46 | Not recorded | 68fa62bf1e209131ce9fb1ba |
| WaveNet | 4.21 | Not recorded | 68fa62bf1e209131ce9fb1ba |
| HMM-driven concatenative | 3.86 | Not recorded | 68fa62bf1e209131ce9fb1ba |
| LSTM-RNN parametric | 3.67 | Not recorded | 68fa62bf1e209131ce9fb1ba |

### Natural

MOS: 4.55. Result ID: 698be200ecace47ef5599018.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62bf1e209131ce9fb1ba.md>)

Recorded evaluation configuration: {"Encoding":"16-bit linear PCM"}. This record does not assert independent reproduction.

### Natural

MOS: 4.46. Result ID: 698be200ecace47ef5599017.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62bf1e209131ce9fb1ba.md>)

Recorded evaluation configuration: {"Encoding":"8-bit mu-law"}. This record does not assert independent reproduction.

### WaveNet

MOS: 4.21. Result ID: 698be200ecace47ef5599016.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62bf1e209131ce9fb1ba.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62bf1e209131ce9fb1ba.md>)

Recorded evaluation configuration: {"Conditioning":"L+F (Linguistic features + log F0)"}. This record does not assert independent reproduction.

### HMM-driven concatenative

MOS: 3.86. Result ID: 698be200ecace47ef5599015.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62bf1e209131ce9fb1ba.md>)

Recorded evaluation configuration: {"Architecture":"Unit selection concatenative"}. This record does not assert independent reproduction.

### LSTM-RNN parametric

MOS: 3.67. Result ID: 698be200ecace47ef5599014.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62bf1e209131ce9fb1ba.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa629d1e209131ce9fa36c.md>)

Recorded evaluation configuration: {"Architecture":"Statistical parametric"}. This record does not assert independent reproduction.

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