Upload README.md with huggingface_hub

README.md

@@ -1,21 +1,67 @@
 ---
 language: [ne]
 license: cc-by-nc-4.0
-tags: [automatic-speech-recognition, nemo, conformer, telephony, nepali]
+library_name: nemo
+pipeline_tag: automatic-speech-recognition
+tags: [automatic-speech-recognition, speech, nemo, conformer, streaming, telephony, nepali, nepal]
+model-index:
+- name: nepali-conformer-streaming
+  results:
+  - task:
+      type: automatic-speech-recognition
+    dataset:
+      name: NepTel v0.1 (real Nepali call-center audio, human-reviewed)
+      type: neptel
+    metrics:
+    - type: wer
+      value: 59.87
+      name: Real-call WER
+    - type: cer
+      value: 41.08
+      name: Real-call CER
+  - task:
+      type: automatic-speech-recognition
+    dataset:
+      name: Held-out gold read Nepali (W1 read slice, OpenSLR-54 utterances absent from training)
+      type: w1-read
+    metrics:
+    - type: wer
+      value: 31.5
+      name: Read-speech WER
 ---
 
+<p align="center"><img src="https://raw.githubusercontent.com/Ampixa/nepaliconformer/master/assets/logo.svg" width="140"></p>
+
 # nepali-conformer-streaming
 
-Cache-aware streaming Nepali ASR (520 ms lookahead, chunked attention, causal convs). Carries a large measured streaming-lineage penalty — see RESULTS.md before using.
+Cache-aware **streaming** Nepali ASR (520 ms lookahead). Carries a large, honestly-reported streaming-lineage penalty on real calls — read RESULTS.md before choosing this over the offline model; it exists because a phone agent needs incremental output.
+
+**Try it:** [demo Space](https://huggingface.co/spaces/voidash/nepaliconformer) ·
+**Everything else:** [github.com/Ampixa/nepaliconformer](https://github.com/Ampixa/nepaliconformer)
+(NepTel benchmark, per-system outputs, full honest results)
+
+## Numbers (measured, not marketed)
+
+| benchmark | WER |
+|---|---|
+| NepTel — real Nepali call audio, human-reviewed refs | **59.87** |
+| Held-out gold read Nepali (W1 slice) | 31.5 |
+| Whisper-large-v3 zero-shot on the same NepTel audio | 99.4 |
+
+## Architecture
+
+121.3M-parameter 17-layer Conformer (d=512, striding ×4, 40 ms frames), hybrid TDT/CTC decoder,
+1,024-piece Devanagari SentencePiece. Chunked-limited attention [[70,13],[70,6],[70,1],[70,0]], fully causal convolutions, cache-aware incremental decoding.
+
+## Training data
 
-Part of the [nepali-telephony-asr](https://github.com/voidash/nepali-telephony-asr) research
-prototype release. **Read the honest results first**: real-call WER 59.9 on the NepTel
-benchmark (human-reviewed real Nepali call audio), with measured limitations (English, sung
-speech, slow speech, end-of-turn) documented in the repo's RESULTS.md.
+~1,655 h of mostly conversational Nepali (YouTube podcasts/interviews) with Google Chirp 2
+pseudo-labels + 105 h human-labeled read speech; telephony codec, noise, reverb and tempo
+augmentation. Label-noise ceiling and every measured limitation (English, sung speech, slow
+speech, end-of-turn) are documented in the
+[repo's RESULTS.md](https://github.com/Ampixa/nepaliconformer/blob/master/RESULTS.md).
 
-121.3M-parameter 17-layer Conformer (d=512, striding x4, 40 ms frames), hybrid TDT/CTC,
-1024-piece SentencePiece (Devanagari). Trained from scratch on ~1,655 h of mostly conversational
-Nepali with Chirp 2 pseudo-labels; telephony codec + noise + tempo augmentation.
+## Usage
 
 ```python
 from nemo.collections.asr.models import EncDecHybridRNNTCTCBPEModel
@@ -23,4 +69,4 @@ m = EncDecHybridRNNTCTCBPEModel.restore_from("nepali_conformer_streaming.nemo")
 print(m.transcribe(["audio.wav"])[0].text)
 ```
 
-License: CC-BY-NC-4.0 (weights; training data includes crawled YouTube speech).
+License: CC-BY-NC-4.0 (weights). Code in the repo: MIT.