# Automatic Speech Recognition on MCIF constrained, short audio

Source: [Automatic Speech Recognition on MCIF constrained, short audio](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-mcif-constrained-short-audio>)

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

Data updated: 2026-07-08T06:55:18.572Z

Evaluation of Automatic Speech Recognition performance on the MCIF constrained, short audio dataset, measured using Word Error Rate (WER).

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) | MCIF constrained, short audio |
| Metric | 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

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

## Recorded results

| Method | WER | Result date | Report paper |
| --- | --- | --- | --- |
| SeamlessM4T-v2-large (base) | 21.49 | Not recorded | 6a4dd9ad4da6617fe9198bee |
| Stage 3 (A.1 + V1 LoRA) | 23.49 | Not recorded | 6a4dd9ad4da6617fe9198bee |
| A.1 Projector (ASR/ST) | 28.9 | Not recorded | 6a4dd9ad4da6617fe9198bee |
| A.2 Projector (ASR/ST/SQA) | 37.94 | Not recorded | 6a4dd9ad4da6617fe9198bee |

### SeamlessM4T-v2-large (base)

WER: 21.49. Result ID: 6a4df44b60b88bb339d69547.

[Reporting paper](<https://www.sota2.com/research/paper/6a4dd9ad4da6617fe9198bee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70f21e209131cea4e39a.md>)

Recorded evaluation configuration: {"status":"base"}. This record does not assert independent reproduction.

### Stage 3 (A.1 + V1 LoRA)

WER: 23.49. Result ID: 6a4df44b60b88bb339d6954a.

[Reporting paper](<https://www.sota2.com/research/paper/6a4dd9ad4da6617fe9198bee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a4dd9ad4da6617fe9198bee.md>)

Recorded evaluation configuration: {"Stage":"Stage 3","Components":"A.1 + V1 LoRA"}. This record does not assert independent reproduction.

### A.1 Projector (ASR/ST)

WER: 28.9. Result ID: 6a4df44b60b88bb339d69548.

[Reporting paper](<https://www.sota2.com/research/paper/6a4dd9ad4da6617fe9198bee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a4dd9ad4da6617fe9198bee.md>)

Recorded evaluation configuration: {"Stage":"Stage 1","Modules":"Projector"}. This record does not assert independent reproduction.

### A.2 Projector (ASR/ST/SQA)

WER: 37.94. Result ID: 6a4df44b60b88bb339d69549.

[Reporting paper](<https://www.sota2.com/research/paper/6a4dd9ad4da6617fe9198bee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a4dd9ad4da6617fe9198bee.md>)

Recorded evaluation configuration: {"Stage":"Stage 1","Modules":"Projector"}. This record does not assert independent reproduction.

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