initial commit

app.py

@@ -0,0 +1,54 @@
+import gradio as gr
+from transformers import AutoFeatureExtractor, Wav2Vec2BertModel
+import torch
+
+MODEL_NAME = "mikr/w2v-bert-2.0-czech-colab-cv16"
+lang = "cs"
+
+device = 0 if torch.cuda.is_available() else "cpu"
+pipe = pipeline(
+    task="automatic-speech-recognition",
+    model=MODEL_NAME,
+    chunk_length_s=30,
+    device=device,
+)
+
+pipe.model.config.forced_decoder_ids = pipe.tokenizer.get_decoder_prompt_ids(language=lang, task="transcribe")
+
+def transcribe(microphone, file_upload):
+    warn_output = ""
+    if (microphone is not None) and (file_upload is not None):
+        warn_output = (
+            "WARNING: You've uploaded an audio file and used the microphone. "
+            "The recorded file from the microphone will be used and the uploaded audio will be discarded.\n"
+        )
+
+    elif (microphone is None) and (file_upload is None):
+        return "ERROR: You have to either use the microphone or upload an audio file"
+
+    file = microphone if microphone is not None else file_upload
+
+    text = pipe(file)["text"]
+
+    return warn_output + text
+
+
+iface = gr.Interface(
+    fn=transcribe,
+    inputs=[
+        gr.inputs.Audio(source="microphone", type="filepath", optional=True),
+        gr.inputs.Audio(source="upload", type="filepath", optional=True),
+    ],
+    outputs="text",
+    layout="horizontal",
+    theme="huggingface",
+    title="Wav2Vec2-Bert demo - transcribe Czech Audio",
+    description=(
+        "Transcribe long-form microphone or audio inputs with the click of a button! Demo uses the fine-tuned"
+        f" checkpoint [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) from Whisper Fine Tuning Sprint Event 2022 "
+        "and 🤗 Transformers to transcribe audio files of arbitrary length."
+    ),
+    allow_flagging="never",
+)
+
+iface.launch()