Spaces:
Running
Running
| title: Nawah ASR 50M | |
| emoji: 🎧 | |
| colorFrom: blue | |
| colorTo: indigo | |
| sdk: docker | |
| app_port: 7860 | |
| pinned: false | |
| short_description: تفريغ صوتي عربي بموديل Emhotob 50M | |
| # Nawah-ASR-50M | |
| Arabic ASR where a **50M Arabic LLM writes the transcript**. A frozen Whisper-small encoder feeds | |
| a 0.39M projector; [`oddadmix/50M-2048-Emhotob`](https://huggingface.co/oddadmix/50M-2048-Emhotob) | |
| does the generating. | |
| Record or upload a clip (up to 30s). The demo reports the **audio token count**, which tracks the | |
| clip's real duration at 25 tokens/sec -- a 3-second clip costs 75 tokens, not the 375 a fixed | |
| 30-second window would spend on silence. | |
| Scored on a 3,165-clip video-disjoint MASC split (same clips, same normalizer for every row): | |
| | model | decoder | WER | CER | | |
| |---|---|---|---| | |
| | `openai/whisper-small` (zero-shot) | Whisper 153M | 0.5739 | 0.2062 | | |
| | `whisper-small-arabic-dialectal` | Whisper 153M | 0.6775 | 0.2244 | | |
| | `Nawah-ASR-50M-v1` (60 h) | Emhotob 50M | 0.6145 | 0.2936 | | |
| | **`Nawah-ASR-50M-v1.1` (300 h)** | **Emhotob 50M** | **0.4109** | **0.1815** | | |
| With the encoder fixed, swapping Whisper's 153M decoder for Emhotob's 50M takes WER 0.5739 -> | |
| 0.4109. Caveat: this is MASC's home turf and both Whisper checkpoints are out-of-domain here -- | |
| the tell is that the dialect-tuned one scores *worse* than the base model on this MSA-leaning data. | |
| Model: [`oddadmix/Nawah-ASR-50M-v1.1`](https://huggingface.co/oddadmix/Nawah-ASR-50M-v1.1) | |