fix readme.md

README.md

@@ -1,22 +1,83 @@
 ---
 pipeline_tag: automatic-speech-recognition
 license: apache-2.0
+base_model: openai/whisper-medium
 library_name: kerasformers
 tags:
 - keras
 - kerasformers
 - whisper
-- tf
-- jax
+- automatic-speech-recognition
+- audio
+- multilingual
+- arxiv:2212.04356
 - pytorch
+- jax
+- tf
 ---
 
-# whisper_medium (Keras 3)
+## ***See [our collection](https://huggingface.co/collections/kerasformers/whisper-6a6ac856710a03d7d2e207b6) for all versions of Whisper.***
+
+# Run Whisper with Keras 3: JAX, PyTorch, or TensorFlow
+
+[![GitHub](https://img.shields.io/badge/GitHub-KerasFormers-black?logo=github)](https://github.com/IMvision12/KerasFormers) [![Docs](https://img.shields.io/badge/Docs-Whisper-blue)](https://imvision12.github.io/KerasFormers/whisper/) [![Collection](https://img.shields.io/badge/HF-Whisper%20collection-yellow)](https://huggingface.co/collections/kerasformers/whisper-6a6ac856710a03d7d2e207b6)
+
+# kerasformers/whisper_medium
+
+Paper: [Robust Speech Recognition via Large-Scale Weak Supervision (arXiv:2212.04356)](https://arxiv.org/abs/2212.04356) · [HF Papers](https://huggingface.co/papers/2212.04356)
+
+Whisper is a **multilingual** encoder-decoder ASR model trained on large-scale weak supervision. Use `task="transcribe"` to keep the source language or `task="translate"` to render English. Pass `language=None` to let the model detect the spoken language. Output is cased and punctuated.
+
+For more details on the model, please go to the upstream [model card](https://huggingface.co/openai/whisper-medium).
+
+Pure-**Keras 3** conversion of [`openai/whisper-medium`](https://huggingface.co/openai/whisper-medium) for [kerasformers](https://github.com/IMvision12/KerasFormers). One implementation runs unmodified on **TensorFlow / Torch / JAX**.
 
-Pure-Keras 3 weights for [kerasformers](https://github.com/IMvision12/KerasFormers), mirrored from the GitHub release. License: `apache-2.0`.
+This is an **ASR** checkpoint (`WhisperSpeechToText`, 769M).
+
+## ✨ Quick start
 
 ```python
-from kerasformers.models.whisper import WhisperSpeechToText, WhisperProcessor
-model = WhisperSpeechToText.from_weights("whisper_medium")
-processor = WhisperProcessor.from_weights("whisper_medium")
+import os
+os.environ["KERAS_BACKEND"] = "torch"  # or "jax" / "tensorflow"
+
+import soundfile as sf
+from kerasformers.models.whisper import (
+    WhisperProcessor,
+    WhisperSpeechToText,
+)
+
+model = WhisperSpeechToText.from_weights("kerasformers/whisper_medium")
+processor = WhisperProcessor.from_weights("kerasformers/whisper_medium")
+
+audio, sr = sf.read("your_audio.wav", dtype="float32")  # 16 kHz mono
+# task="transcribe" keeps the source language; "translate" -> English.
+text = model.generate(audio, processor, language="en", task="transcribe")
+print(repr(text[0]))
 ```
+
+Load any Whisper variant the same way with `from_weights("kerasformers/<variant>")`:
+
+| Variant | Hub | Notes |
+|---|---|---|
+| `whisper_tiny` | [`kerasformers/whisper_tiny`](https://huggingface.co/kerasformers/whisper_tiny) | 39M |
+| `whisper_base` | [`kerasformers/whisper_base`](https://huggingface.co/kerasformers/whisper_base) | 74M |
+| `whisper_small` | [`kerasformers/whisper_small`](https://huggingface.co/kerasformers/whisper_small) | 244M |
+| `whisper_medium` | [`kerasformers/whisper_medium`](https://huggingface.co/kerasformers/whisper_medium) | 769M |
+| `whisper_large` | [`kerasformers/whisper_large`](https://huggingface.co/kerasformers/whisper_large) | 1.55B |
+| `whisper_large_v2` | [`kerasformers/whisper_large_v2`](https://huggingface.co/kerasformers/whisper_large_v2) | 1.55B |
+| `whisper_large_v3` | [`kerasformers/whisper_large_v3`](https://huggingface.co/kerasformers/whisper_large_v3) | 128 mel bins |
+| `whisper_large_v3_turbo` | [`kerasformers/whisper_large_v3_turbo`](https://huggingface.co/kerasformers/whisper_large_v3_turbo) | 4 decoder layers |
+
+## Tips
+
+- Set `KERAS_BACKEND` **before** importing Keras / kerasformers.
+- Prefer `WhisperProcessor.from_weights(...)` so mel bins match the variant (v3 uses 128).
+- Clips are padded to a 30 s window; chunk longer audio yourself.
+- See [Whisper docs](https://imvision12.github.io/KerasFormers/whisper/) and [Loading Weights](https://imvision12.github.io/KerasFormers/loading_weights/).
+- Community / upstream safetensors still work via the `hf:` prefix, e.g. `WhisperSpeechToText.from_weights("hf:openai/whisper-medium")`.
+
+## Special Thanks
+
+A huge thank you to the OpenAI Whisper authors for creating and releasing these models.
+
+License: Apache 2.0.