# Speech-to-SQL Parsing on MASpider Custom Data generalization (test)

Source: [Speech-to-SQL Parsing on MASpider Custom Data generalization (test)](<https://www.sota2.com/research/sota/speech-to-sql-parsing-on-maspider-custom-data-generalization-test>)

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

Data updated: 2026-02-26T06:02:52.803Z

Evaluation of Speech-to-SQL performance on the test split of the MASpider Custom Data generalization dataset, measured by structural SQL components (SELECT, WHERE, GROUP, ORDER) and Exact Match.

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 | Speech-to-SQL Parsing |
| Dataset (including any stated split) | MASpider Custom Data generalization (test) |
| Metric | SELECT Component Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [SELECT Component Accuracy](<https://www.sota2.com/research/sota/speech-to-sql-parsing-on-maspider-custom-data-generalization-test.md?metric=SELECT%20Component%20Accuracy>): Higher is better; unit: PERCENT.

- [WHERE Component Accuracy](<https://www.sota2.com/research/sota/speech-to-sql-parsing-on-maspider-custom-data-generalization-test.md?metric=WHERE%20Component%20Accuracy>): Higher is better; unit: PERCENT.

- [GROUP BY Component Accuracy](<https://www.sota2.com/research/sota/speech-to-sql-parsing-on-maspider-custom-data-generalization-test.md?metric=GROUP%20BY%20Component%20Accuracy>): Higher is better; unit: PERCENT.

- [ORDER BY Component Accuracy](<https://www.sota2.com/research/sota/speech-to-sql-parsing-on-maspider-custom-data-generalization-test.md?metric=ORDER%20BY%20Component%20Accuracy>): Higher is better; unit: PERCENT.

- [Logical Operator Accuracy (AND/OR)](<https://www.sota2.com/research/sota/speech-to-sql-parsing-on-maspider-custom-data-generalization-test.md?metric=Logical%20Operator%20Accuracy%20(AND%2FOR)>): Higher is better; unit: PERCENT.

- [Keyword Accuracy](<https://www.sota2.com/research/sota/speech-to-sql-parsing-on-maspider-custom-data-generalization-test.md?metric=Keyword%20Accuracy>): Higher is better; unit: PERCENT.

- [Exact Match Accuracy](<https://www.sota2.com/research/sota/speech-to-sql-parsing-on-maspider-custom-data-generalization-test.md?metric=Exact%20Match%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | SELECT Component Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| S2SQL-TTS | 67.8 | Not recorded | 68fa6f641e209131cea4543d |
| Cascaded | 61.9 | Not recorded | 68fa6f641e209131cea4543d |
| Wav2SQL | 61.1 | Not recorded | 68fa6f641e209131cea4543d |
| DeepSpeechSQL | 41 | Not recorded | 68fa6f641e209131cea4543d |

### S2SQL-TTS

SELECT Component Accuracy: 67.8. Result ID: 6990ba283a6c31484902c7e3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f641e209131cea4543d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d351e209131cea386ad.md>)

[Implementation](<https://speechresearch.github.io/fastspeech2/>)

Recorded evaluation configuration: {"Synthesized data upper bound":"FastSpeech 2"}. This record does not assert independent reproduction.

### Cascaded

SELECT Component Accuracy: 61.9. Result ID: 6990ba283a6c31484902c7e5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f641e209131cea4543d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68891e209131cea1cc36.md>)

[Implementation](<https://github.com/pytorch/fairseq>)

Recorded evaluation configuration: {"ASR":"wav2vec 2.0","Text-to-SQL parser":"RAT-SQL"}. This record does not assert independent reproduction.

### Wav2SQL

SELECT Component Accuracy: 61.1. Result ID: 6990ba283a6c31484902c7e6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f641e209131cea4543d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f641e209131cea4543d.md>)

[Implementation](<https://wav2sql.github.io/>)

Recorded evaluation configuration: {"Speech feature extractor":"Hubert","Language model":"GloVe"}. This record does not assert independent reproduction.

### DeepSpeechSQL

SELECT Component Accuracy: 41. Result ID: 6990ba283a6c31484902c7e4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f641e209131cea4543d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62321e209131ce9f82cd.md>)

Recorded evaluation configuration: {"Speech encoder":"DeepSpeech"}. This record does not assert independent reproduction.

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