# Automatic Speech Recognition on MS-SNSD SNR = 10 dB (test)

Source: [Automatic Speech Recognition on MS-SNSD SNR = 10 dB (test)](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-ms-snsd-snr-10-db-test>)

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

Data updated: 2026-03-06T09:33:12.070Z

Evaluation of Automatic Speech Recognition performance on the test split of the MS-SNSD dataset under 10 dB Signal-to-Noise Ratio (SNR) conditions, measured using various segmented Word Error Rate (WER) metrics and latency.

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) | MS-SNSD SNR = 10 dB (test) |
| Metric | AC WER |
| Unit | PERCENT |
| Direction | Lower 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

- [AC WER](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-ms-snsd-snr-10-db-test.md?metric=AC%20WER>): Lower is better; unit: PERCENT.

- [AA WER](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-ms-snsd-snr-10-db-test.md?metric=AA%20WER>): Lower is better; unit: PERCENT.

- [BA WER](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-ms-snsd-snr-10-db-test.md?metric=BA%20WER>): Lower is better; unit: PERCENT.

- [CM WER](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-ms-snsd-snr-10-db-test.md?metric=CM%20WER>): Lower is better; unit: PERCENT.

- [MU WER](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-ms-snsd-snr-10-db-test.md?metric=MU%20WER>): Lower is better; unit: PERCENT.

- [NB WER](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-ms-snsd-snr-10-db-test.md?metric=NB%20WER>): Lower is better; unit: PERCENT.

- [SD WER](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-ms-snsd-snr-10-db-test.md?metric=SD%20WER>): Lower is better; unit: PERCENT.

- [TP WER](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-ms-snsd-snr-10-db-test.md?metric=TP%20WER>): Lower is better; unit: PERCENT.

- [Mean WER](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-ms-snsd-snr-10-db-test.md?metric=Mean%20WER>): Lower is better; unit: PERCENT.

- [Latency (s)](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-ms-snsd-snr-10-db-test.md?metric=Latency%20(s)>): Lower is better; unit: TIME.

## Recorded results

| Method | AC WER | Result date | Report paper |
| --- | --- | --- | --- |
| ASR-TRA | 22.39 | Not recorded | 69aa5f932d603e6de3e4198a |
| SGEM | 23.76 | Not recorded | 69aa5f932d603e6de3e4198a |
| SUTA | 24.02 | Not recorded | 69aa5f932d603e6de3e4198a |
| Baseline | 26.54 | Not recorded | 69aa5f932d603e6de3e4198a |

### ASR-TRA

AC WER: 22.39. Result ID: 69aa9f4e6dbb7075b954affd.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f932d603e6de3e4198a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69aa5f932d603e6de3e4198a.md>)

Recorded evaluation configuration: {"Model":"Whisper-Tiny","Adaptation Method":"ASR-TRA"}. This record does not assert independent reproduction.

### SGEM

AC WER: 23.76. Result ID: 69aa9f4e6dbb7075b954affc.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f932d603e6de3e4198a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fc61e209131cea4756e.md>)

Recorded evaluation configuration: {"Model":"Whisper-Tiny","Adaptation Method":"SGEM"}. This record does not assert independent reproduction.

### SUTA

AC WER: 24.02. Result ID: 69aa9f4e6dbb7075b954affb.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f932d603e6de3e4198a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cda1e209131cea3644f.md>)

Recorded evaluation configuration: {"Model":"Whisper-Tiny","Adaptation Method":"SUTA"}. This record does not assert independent reproduction.

### Baseline

AC WER: 26.54. Result ID: 69aa9f4e6dbb7075b954affa.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f932d603e6de3e4198a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e2c1e209131cea3e51b.md>)

[Implementation](<https://github.com/openai/whisper>)

Recorded evaluation configuration: {"Model":"Whisper-Tiny"}. This record does not assert independent reproduction.

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