# Automatic Speech Recognition on Valibel corpus (test)

Source: [Automatic Speech Recognition on Valibel corpus (test)](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-valibel-corpus-test>)

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

Data updated: 2026-02-26T06:08:27.876Z

Evaluation of Automatic Speech Recognition performance on the test split of the Valibel corpus, typically measured using Word Error Rate (WER) and Character Error Rate (CER).

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 | Automatic Speech Recognition |
| Dataset (including any stated split) | Valibel corpus (test) |
| Metric | WER |
| Unit | PERCENT |
| Direction | Lower is better |
| Results loaded | 10 of 10 |
| 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

- [WER](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-valibel-corpus-test.md?metric=WER>): Lower is better; unit: PERCENT.

- [CER](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-valibel-corpus-test.md?metric=CER>): Lower is better; unit: PERCENT.

## Recorded results

| Method | WER | Result date | Report paper |
| --- | --- | --- | --- |
| TEXT SEQ2LABEL | 0 | Not recorded | 68fa73ab1e209131cea5dceb |
| TEXT GRAPH | 0 | Not recorded | 68fa73ab1e209131cea5dceb |
| TEXT HOPS | 0 | Not recorded | 68fa73ab1e209131cea5dceb |
| PIPELINE SEQ2LABEL | 30.5 | Not recorded | 68fa73ab1e209131cea5dceb |
| PIPELINE GRAPH | 30.5 | Not recorded | 68fa73ab1e209131cea5dceb |
| PIPELINE HOPS | 30.5 | Not recorded | 68fa73ab1e209131cea5dceb |
| AUDIO GRAPH | 30.6 | Not recorded | 68fa73ab1e209131cea5dceb |
| ORACLE SEQ2LABEL | 30.9 | Not recorded | 68fa73ab1e209131cea5dceb |
| AUDIO SEQ2LABEL | 31 | Not recorded | 68fa73ab1e209131cea5dceb |
| ORACLE GRAPH | 31.4 | Not recorded | 68fa73ab1e209131cea5dceb |

### TEXT SEQ2LABEL

WER: 0. Result ID: 699307663a6c314849132cc3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ab1e209131cea5dceb.md>)

Recorded evaluation configuration: {"Input":"Text (Gold)","Parameters":"110M + 39.2M","Pre-training":"CamemBERT"}. This record does not assert independent reproduction.

### TEXT GRAPH

WER: 0. Result ID: 699307663a6c314849132cc4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ab1e209131cea5dceb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73ab1e209131cea5dceb.md>)

[Implementation](<https://github.com/Pupiera/Growing_tree_on_sound>)

Recorded evaluation configuration: {"Input":"Text (Gold)","Parameters":"110M + 41.4M","Pre-training":"CamemBERT"}. This record does not assert independent reproduction.

### TEXT HOPS

WER: 0. Result ID: 699307663a6c314849132cc5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ab1e209131cea5dceb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73ab1e209131cea5dceb.md>)

[Implementation](<https://github.com/Pupiera/Growing_tree_on_sound>)

Recorded evaluation configuration: {"Input":"Text (Gold)","Parameters":"110M + 100M","Pre-training":"CamemBERT"}. This record does not assert independent reproduction.

### PIPELINE SEQ2LABEL

WER: 30.5. Result ID: 699307663a6c314849132cc0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ab1e209131cea5dceb.md>)

Recorded evaluation configuration: {"Input":"Pipeline","Parameters":"314M + 110M + 39.2M","Pre-training":"Wav2vec2 + CamemBERT"}. This record does not assert independent reproduction.

### PIPELINE GRAPH

WER: 30.5. Result ID: 699307663a6c314849132cc1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ab1e209131cea5dceb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73ab1e209131cea5dceb.md>)

[Implementation](<https://github.com/Pupiera/Growing_tree_on_sound>)

Recorded evaluation configuration: {"Input":"Pipeline","Parameters":"314M + 110M + 41.4M","Pre-training":"Wav2vec2 + CamemBERT"}. This record does not assert independent reproduction.

### PIPELINE HOPS

WER: 30.5. Result ID: 699307663a6c314849132cc2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ab1e209131cea5dceb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73ab1e209131cea5dceb.md>)

[Implementation](<https://github.com/Pupiera/Growing_tree_on_sound>)

Recorded evaluation configuration: {"Input":"Pipeline","Parameters":"314M + 110M + 100M","Pre-training":"Wav2vec2 + CamemBERT"}. This record does not assert independent reproduction.

### AUDIO GRAPH

WER: 30.6. Result ID: 699307663a6c314849132cbd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ab1e209131cea5dceb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73ab1e209131cea5dceb.md>)

[Implementation](<https://github.com/Pupiera/Growing_tree_on_sound>)

Recorded evaluation configuration: {"Input":"Audio","Parameters":"315M + 34.9M","Pre-training":"Wav2vec2"}. This record does not assert independent reproduction.

### ORACLE SEQ2LABEL

WER: 30.9. Result ID: 699307663a6c314849132cbe.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ab1e209131cea5dceb.md>)

Recorded evaluation configuration: {"Input":"Oracle","Parameters":"315M + 34.9M","Pre-training":"Wav2vec2"}. This record does not assert independent reproduction.

### AUDIO SEQ2LABEL

WER: 31. Result ID: 699307663a6c314849132cbc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ab1e209131cea5dceb.md>)

Recorded evaluation configuration: {"Input":"Audio","Parameters":"315M + 34.9M","Pre-training":"Wav2vec2"}. This record does not assert independent reproduction.

### ORACLE GRAPH

WER: 31.4. Result ID: 699307663a6c314849132cbf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73ab1e209131cea5dceb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73ab1e209131cea5dceb.md>)

[Implementation](<https://github.com/Pupiera/Growing_tree_on_sound>)

Recorded evaluation configuration: {"Input":"Oracle","Parameters":"315M + 34.9M","Pre-training":"Wav2vec2"}. This record does not assert independent reproduction.

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