# Pitch Reconstruction on LibriTTS (test)

Source: [Pitch Reconstruction on LibriTTS (test)](<https://www.sota2.com/research/sota/pitch-reconstruction-on-libritts-test>)

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

Data updated: 2026-06-02T08:07:59.632Z

Evaluation of pitch reconstruction quality on the test split of the LibriTTS dataset, measured using Mean Squared Error (MSE) based metrics.

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 | Pitch reconstruction |
| Dataset (including any stated split) | LibriTTS (test) |
| Metric | Standard MSE |
| Unit | SCALAR |
| Direction | Lower 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

- [Standard MSE](<https://www.sota2.com/research/sota/pitch-reconstruction-on-libritts-test.md?metric=Standard%20MSE>): Lower is better; unit: SCALAR.

- [0-Mean MSE](<https://www.sota2.com/research/sota/pitch-reconstruction-on-libritts-test.md?metric=0-Mean%20MSE>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Standard MSE | Result date | Report paper |
| --- | --- | --- | --- |
| Prosody Encoder | 0.024 | Not recorded | 6a1e63b9dbd7150506f852e1 |
| Prosody Encoder | 0.025 | Not recorded | 6a1e63b9dbd7150506f852e1 |
| Prosody Encoder | 0.027 | Not recorded | 6a1e63b9dbd7150506f852e1 |
| Prosody Encoder | 0.048 | Not recorded | 6a1e63b9dbd7150506f852e1 |
| HuBERT-base | 0.056 | Not recorded | 6a1e63b9dbd7150506f852e1 |

### Prosody Encoder

Standard MSE: 0.024. Result ID: 6a1e8f54fbc373592c1890eb.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b9dbd7150506f852e1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a1e63b9dbd7150506f852e1.md>)

Recorded evaluation configuration: {"Speaker-norm. log F0":"false","Adversarial speaker loss (L\_adv\_spk)":"true"}. This record does not assert independent reproduction.

### Prosody Encoder

Standard MSE: 0.025. Result ID: 6a1e8f54fbc373592c1890ec.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b9dbd7150506f852e1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a1e63b9dbd7150506f852e1.md>)

Recorded evaluation configuration: {"Speaker-norm. log F0":"true","Adversarial speaker loss (L\_adv\_spk)":"true"}. This record does not assert independent reproduction.

### Prosody Encoder

Standard MSE: 0.027. Result ID: 6a1e8f54fbc373592c1890e9.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b9dbd7150506f852e1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a1e63b9dbd7150506f852e1.md>)

Recorded evaluation configuration: {"Speaker-norm. log F0":"false","Adversarial speaker loss (L\_adv\_spk)":"false"}. This record does not assert independent reproduction.

### Prosody Encoder

Standard MSE: 0.048. Result ID: 6a1e8f54fbc373592c1890ea.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b9dbd7150506f852e1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a1e63b9dbd7150506f852e1.md>)

Recorded evaluation configuration: {"Speaker-norm. log F0":"true","Adversarial speaker loss (L\_adv\_spk)":"false"}. This record does not assert independent reproduction.

### HuBERT-base

Standard MSE: 0.056. Result ID: 6a1e8f54fbc373592c1890e8.

[Reporting paper](<https://www.sota2.com/research/paper/6a1e63b9dbd7150506f852e1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a2a1e209131cea26880.md>)

[Implementation](<https://github.com/pytorch/fairseq/tree/master/examples/hubert>)

Recorded evaluation configuration: {"Adversarial speaker loss (L\_adv\_spk)":"false"}. This record does not assert independent reproduction.

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