README.md

---
pretty_name: SaarAI ASR Benchmark Outputs
language:
  - amh
  - kin
  - nde
  - sna
  - zul
task_categories:
  - automatic-speech-recognition
tags:
  - asr
  - benchmark
  - model-outputs
  - saarai
configs:
  - config_name: amharic
    data_files:
      - split: ambient_room_noise
        path: "data/amharic/ambient_room_noise__*.jsonl"
      - split: clean_studio
        path: "data/amharic/clean_studio__*.jsonl"
      - split: distant_low_volume_mic
        path: "data/amharic/distant_low_volume_mic__*.jsonl"
      - split: harsh_environment
        path: "data/amharic/harsh_environment__*.jsonl"
  - config_name: kinyarwanda
    data_files:
      - split: test
        path: "data/kinyarwanda/test__*.jsonl"
  - config_name: ndebele
    data_files:
      - split: test
        path: "data/ndebele/test__*.jsonl"
  - config_name: shona
    data_files:
      - split: test
        path: "data/shona/test__*.jsonl"
  - config_name: zulu
    data_files:
      - split: test
        path: "data/zulu/test__*.jsonl"
---

# SaarAI ASR Benchmark Outputs

Raw per-utterance model outputs (transcription manifests) produced by the [gsma-asr-bench](https://github.com/IsraelAbebe/gsma-asr-bench) runners on `SaarAI/asr-leaderboard-datasets`.

- files: **84**
- utterances: **373416**
- languages: **5**
- models: **22**

## Layout

```
data/<language_name>/<split>__<dataset_config>__<model_slug>.jsonl
index.jsonl   # one record per file (language, split, model, rows, sha256, ...)
index.csv
```

Directories categorise by **language name**; the file name begins with the **split name**, so both can be parsed without opening the file:

```python
from pathlib import Path
split, dataset_config, model_slug = Path(path).stem.split('__', 2)
language_name = Path(path).parent.name
```

## Loading

```python
from datasets import load_dataset

# everything
ds = load_dataset("SaarAI/asr-benchmark-outputs", data_files="data/**/*.jsonl", split="train")

# one language / one split
ds = load_dataset("SaarAI/asr-benchmark-outputs", data_files="data/amharic/ambient_room_noise__*.jsonl", split="train")
```

## Row schema

`audio_filepath`, `duration`, `time`, `text` (reference), `pred_text` (hypothesis) plus the metadata added at upload time: `model_slug`, `model_family`, `language`, `language_name`, `iso_639_3`, `dataset_config`, `split`, `benchmark_dataset`.

## Contents

| language | splits | files | utterances |
| --- | --- | --- | --- |
| amharic | ambient_room_noise, clean_studio, distant_low_volume_mic, harsh_environment | 10 | 159650 |
| kinyarwanda | test | 10 | 100760 |
| ndebele | test | 10 | 19330 |
| shona | test | 13 | 22737 |
| zulu | test | 41 | 70939 |