# Spoken Question Answering on Spoken SQuAD (B) ASR (test)

Source: [Spoken Question Answering on Spoken SQuAD (B) ASR (test)](<https://www.sota2.com/research/sota/spoken-question-answering-on-spoken-squad-b-asr-test>)

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

Data updated: 2026-02-26T05:41:32.592Z

Evaluation of model performance on the test split of the Spoken SQuAD (B) ASR dataset for Spoken Question Answering, reporting EM and F1 scores.

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 | Spoken Question Answering |
| Dataset (including any stated split) | Spoken SQuAD (B) ASR (test) |
| Metric | EM |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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

- [EM](<https://www.sota2.com/research/sota/spoken-question-answering-on-spoken-squad-b-asr-test.md?metric=EM>): Higher is better; unit: PERCENT.

- [F1](<https://www.sota2.com/research/sota/spoken-question-answering-on-spoken-squad-b-asr-test.md?metric=F1>): Higher is better; unit: PERCENT.

## Recorded results

| Method | EM | Result date | Report paper |
| --- | --- | --- | --- |
| Ensembled \[(e) plus (d)\] | 60.37 | Not recorded | 68fa682b1e209131cea1a7b9 |
| Ensembled \[(d) plus (d)\] | 57.88 | Not recorded | 68fa682b1e209131cea1a7b9 |
| BERT | 56.28 | Not recorded | 68fa682b1e209131cea1a7b9 |
| SpeechBERT | 53.42 | Not recorded | 68fa682b1e209131cea1a7b9 |
| BERT | 53.3 | Not recorded | 68fa682b1e209131cea1a7b9 |
| SpeechBERT | 51.19 | Not recorded | 68fa682b1e209131cea1a7b9 |
| SpeechBERT | 46.02 | Not recorded | 68fa682b1e209131cea1a7b9 |
| Dr.QA | 41.16 | Not recorded | 68fa682b1e209131cea1a7b9 |
| BiDAF | 37.02 | Not recorded | 68fa682b1e209131cea1a7b9 |

### Ensembled \[(e) plus (d)\]

EM: 60.37. Result ID: 698d5f148b517eb43320ff37.

[Reporting paper](<https://www.sota2.com/research/paper/68fa682b1e209131cea1a7b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa682b1e209131cea1a7b9.md>)

Recorded evaluation configuration: {"Model Setup":"Ensembled models","Composition":"SpeechBERT + BERT Cascade"}. This record does not assert independent reproduction.

### Ensembled \[(d) plus (d)\]

EM: 57.88. Result ID: 698d5f148b517eb43320ff38.

[Reporting paper](<https://www.sota2.com/research/paper/68fa682b1e209131cea1a7b9.md>)

Recorded evaluation configuration: {"Model Setup":"Ensembled models","Composition":"BERT Cascade + BERT Cascade"}. This record does not assert independent reproduction.

### BERT

EM: 56.28. Result ID: 698d5f148b517eb43320ff33.

[Reporting paper](<https://www.sota2.com/research/paper/68fa682b1e209131cea1a7b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65c31e209131cea0c4ed.md>)

[Implementation](<https://github.com/google-research/bert>)

Recorded evaluation configuration: {"Training set":"ASR \[10\]","Model Setup":"cascade"}. This record does not assert independent reproduction.

### SpeechBERT

EM: 53.42. Result ID: 698d5f148b517eb43320ff36.

[Reporting paper](<https://www.sota2.com/research/paper/68fa682b1e209131cea1a7b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa682b1e209131cea1a7b9.md>)

Recorded evaluation configuration: {"Model Setup":"End-to-end trained on Audio","Boundary Configuration":"better boundaries"}. This record does not assert independent reproduction.

### BERT

EM: 53.3. Result ID: 698d5f148b517eb43320ff32.

[Reporting paper](<https://www.sota2.com/research/paper/68fa682b1e209131cea1a7b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65c31e209131cea0c4ed.md>)

[Implementation](<https://github.com/google-research/bert>)

Recorded evaluation configuration: {"Training set":"Text","Model Setup":"trained on text"}. This record does not assert independent reproduction.

### SpeechBERT

EM: 51.19. Result ID: 698d5f148b517eb43320ff34.

[Reporting paper](<https://www.sota2.com/research/paper/68fa682b1e209131cea1a7b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa682b1e209131cea1a7b9.md>)

Recorded evaluation configuration: {"Model Setup":"End-to-end trained on Audio","Condition":"proposed"}. This record does not assert independent reproduction.

### SpeechBERT

EM: 46.02. Result ID: 698d5f148b517eb43320ff35.

[Reporting paper](<https://www.sota2.com/research/paper/68fa682b1e209131cea1a7b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa682b1e209131cea1a7b9.md>)

Recorded evaluation configuration: {"Model Setup":"End-to-end trained on Audio","Pre-training":"w/o MLM"}. This record does not assert independent reproduction.

### Dr.QA

EM: 41.16. Result ID: 698d5f148b517eb43320ff31.

[Reporting paper](<https://www.sota2.com/research/paper/68fa682b1e209131cea1a7b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa634b1e209131ce9fdcd5.md>)

Recorded evaluation configuration: {"Training set":"Text","Model Setup":"trained on text"}. This record does not assert independent reproduction.

### BiDAF

EM: 37.02. Result ID: 698d5f148b517eb43320ff30.

[Reporting paper](<https://www.sota2.com/research/paper/68fa682b1e209131cea1a7b9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa647d1e209131cea04c91.md>)

[Implementation](<https://www.sota2.com/allenai.github.io/bi-att-flow/>)

Recorded evaluation configuration: {"Training set":"Text","Model Setup":"trained on text"}. This record does not assert independent reproduction.

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