Update app.py

app.py

@@ -1,4 +1,5 @@
 from importlib.metadata import version
+from timeit import default_timer as timer
 
 import gradio as gr
 import numpy as np
@@ -20,46 +21,43 @@ models = {
 }
 
 
-def recoginize(audio: tuple[int, np.ndarray]):
+def recognize(audio: tuple[int, np.ndarray]):
     sample_rate, waveform = audio
     try:
         waveform = waveform.astype(np.float32) / 2 ** (8 * waveform.itemsize - 1)
-        return [[name, model.recognize(waveform, sample_rate=sample_rate, language="ru")] for name, model in models.items()]
+        results = []
+        for name, model in models.items():
+            start = timer()
+            result = model.recognize(waveform, sample_rate=sample_rate, language="ru")
+            time = timer() - start
+            results.append([name, result, f"{time:.3f} s."])
     except Exception as e:
         raise gr.Error(f"{e} Audio: sample_rate: {sample_rate}, waveform.shape: {waveform.shape}.") from e
-
-
-demo = gr.Interface(
-    fn=recoginize,
-    title="ASR demo using onnx-asr (Russian models)",
-    description="""
-    # ONNX ASR
-
-    [![PyPI - Version](https://img.shields.io/pypi/v/onnx-asr.svg)](https://pypi.org/project/onnx-asr)
-    [![PyPI - Downloads](https://img.shields.io/pypi/dm/onnx-asr)](https://pypi.org/project/onnx-asr)
-    [![PyPI - Python Version](https://img.shields.io/pypi/pyversions/onnx-asr.svg)](https://pypi.org/project/onnx-asr)
-    [![PyPI - Types](https://img.shields.io/pypi/types/onnx-asr)](https://pypi.org/project/onnx-asr)
-    [![GitHub License](https://img.shields.io/github/license/istupakov/onnx-asr)](https://github.com/istupakov/onnx-asr/blob/main/LICENSE)
-    [![CI](https://github.com/istupakov/onnx-asr/actions/workflows/python-package.yml/badge.svg)](https://github.com/istupakov/onnx-asr/actions/workflows/python-package.yml)
-
-    **[onnx-asr](https://github.com/istupakov/onnx-asr)** is a Python package for Automatic Speech Recognition using ONNX models. The package is written in pure Python with minimal dependencies (*PyTorch is not required*):
-
-    [![numpy](https://img.shields.io/badge/numpy-required-blue?logo=numpy)](https://pypi.org/project/numpy/)
-    [![onnxruntime](https://img.shields.io/badge/onnxruntime-required-blue?logo=onnx)](https://pypi.org/project/onnxruntime/)
-    [![huggingface-hub](https://img.shields.io/badge/huggingface--hub-optional-blue?logo=huggingface)](https://pypi.org/project/huggingface-hub/)
-
-    ## Models used in demo:
-    * `gigaam-v2-ctc` - Sber GigaAM v2 CTC ([origin](https://github.com/salute-developers/GigaAM), [onnx](https://huggingface.co/istupakov/gigaam-v2-onnx))
-    * `gigaam-v2-rnnt` - Sber GigaAM v2 RNN-T ([origin](https://github.com/salute-developers/GigaAM), [onnx](https://huggingface.co/istupakov/gigaam-v2-onnx))
-    * `nemo-fastconformer-ru-ctc` - Nvidia FastConformer-Hybrid Large (ru) with CTC decoder ([origin](https://huggingface.co/nvidia/stt_ru_fastconformer_hybrid_large_pc), [onnx](https://huggingface.co/istupakov/stt_ru_fastconformer_hybrid_large_pc_onnx))
-    * `nemo-fastconformer-ru-rnnt` - Nvidia FastConformer-Hybrid Large (ru) with RNN-T decoder ([origin](https://huggingface.co/nvidia/stt_ru_fastconformer_hybrid_large_pc), [onnx](https://huggingface.co/istupakov/stt_ru_fastconformer_hybrid_large_pc_onnx))
-    * `alphacep/vosk-model-ru` - Alpha Cephei Vosk 0.54-ru ([origin](https://huggingface.co/alphacep/vosk-model-ru))
-    * `alphacep/vosk-model-small-ru` - Alpha Cephei Vosk 0.52-small-ru ([origin](https://huggingface.co/alphacep/vosk-model-small-ru))
-    * `whisper-base` - OpenAI Whisper Base exported with onnxruntime ([origin](https://huggingface.co/openai/whisper-base), [onnx](https://huggingface.co/istupakov/whisper-base-onnx))
-    """,
-    inputs=[gr.Audio(min_length=1, max_length=20)],
-    outputs=[gr.Dataframe(headers=["Model", "result"], wrap=True, show_fullscreen_button=True)],
-    flagging_mode="never",
-)
+    else:
+        return results
+
+
+with gr.Blocks() as demo:
+    gr.Markdown("""
+    # ASR demo using onnx-asr (Russian models)
+    **[onnx-asr](https://github.com/istupakov/onnx-asr)** is a Python package for Automatic Speech Recognition using ONNX models.
+    The package is written in pure Python with minimal dependencies (no `pytorch` or `transformers`).
+    """)
+    input = gr.Audio(min_length=1, max_length=20)
+    with gr.Row():
+        gr.ClearButton(input)
+        btn = gr.Button("Recognize", variant="primary")
+    output = gr.Dataframe(headers=["model", "result", "time"], wrap=True)
+    btn.click(fn=recognize, inputs=input, outputs=output)
+    with gr.Accordion("ASR models used in this demo", open=False):
+        gr.Markdown("""
+        * `gigaam-v2-ctc` - Sber GigaAM v2 CTC ([origin](https://github.com/salute-developers/GigaAM), [onnx](https://huggingface.co/istupakov/gigaam-v2-onnx))
+        * `gigaam-v2-rnnt` - Sber GigaAM v2 RNN-T ([origin](https://github.com/salute-developers/GigaAM), [onnx](https://huggingface.co/istupakov/gigaam-v2-onnx))
+        * `nemo-fastconformer-ru-ctc` - Nvidia FastConformer-Hybrid Large (ru) with CTC decoder ([origin](https://huggingface.co/nvidia/stt_ru_fastconformer_hybrid_large_pc), [onnx](https://huggingface.co/istupakov/stt_ru_fastconformer_hybrid_large_pc_onnx))
+        * `nemo-fastconformer-ru-rnnt` - Nvidia FastConformer-Hybrid Large (ru) with RNN-T decoder ([origin](https://huggingface.co/nvidia/stt_ru_fastconformer_hybrid_large_pc), [onnx](https://huggingface.co/istupakov/stt_ru_fastconformer_hybrid_large_pc_onnx))
+        * `alphacep/vosk-model-ru` - Alpha Cephei Vosk 0.54-ru ([origin](https://huggingface.co/alphacep/vosk-model-ru))
+        * `alphacep/vosk-model-small-ru` - Alpha Cephei Vosk 0.52-small-ru ([origin](https://huggingface.co/alphacep/vosk-model-small-ru))
+        * `whisper-base` - OpenAI Whisper Base exported with onnxruntime ([origin](https://huggingface.co/openai/whisper-base), [onnx](https://huggingface.co/istupakov/whisper-base-onnx))
+        """)
 
 demo.launch()