Update app.py

app.py

@@ -1,18 +1,15 @@
+import os
+import tempfile
 import torch
-
 import spaces
 import gradio as gr
 import yt_dlp as youtube_dl
+from fastapi import FastAPI, Form, UploadFile, File, HTTPException, Header
+from fastapi.responses import JSONResponse
 from transformers import pipeline
-from transformers.pipelines.audio_utils import ffmpeg_read
-
-import tempfile
-import os
 
 MODEL_NAME = "openai/whisper-large-v3"
 BATCH_SIZE = 8
-FILE_LIMIT_MB = 1000
-YT_LENGTH_LIMIT_S = 3600  # limit to 1 hour YouTube files
 
 device = 0 if torch.cuda.is_available() else "cpu"
 
@@ -23,86 +20,83 @@ pipe = pipeline(
     device=device,
 )
 
+# ---------------------------------------------------------------------------
+# 核心转录逻辑
+# ---------------------------------------------------------------------------
 @spaces.GPU
-def transcribe(inputs, task):
+def run_transcription(audio_bytes_or_path, task: str):
+    return pipe(
+        audio_bytes_or_path,
+        batch_size=BATCH_SIZE,
+        generate_kwargs={"task": task},
+        return_timestamps=True,
+    )["text"]
+
+# Gradio 调用的封装
+def gradio_transcribe(inputs, task):
     if inputs is None:
-        raise gr.Error("No audio file submitted! Please upload or record an audio file before submitting your request.")
-
-    text = pipe(inputs, batch_size=BATCH_SIZE, generate_kwargs={"task": task}, return_timestamps=True)["text"]
-    return  text
-
-
-def _return_yt_html_embed(yt_url):
-    video_id = yt_url.split("?v=")[-1]
-    HTML_str = (
-        f'<center> <iframe width="500" height="320" src="https://www.youtube.com/embed/{video_id}"> </iframe>'
-        " </center>"
-    )
-    return HTML_str
-
-def download_yt_audio(yt_url, filename):
-    info_loader = youtube_dl.YoutubeDL()
-    
-    try:
-        info = info_loader.extract_info(yt_url, download=False)
-    except youtube_dl.utils.DownloadError as err:
-        raise gr.Error(str(err))
-    
-    file_length = info["duration_string"]
-    file_h_m_s = file_length.split(":")
-    file_h_m_s = [int(sub_length) for sub_length in file_h_m_s]
-    
-    if len(file_h_m_s) == 1:
-        file_h_m_s.insert(0, 0)
-    if len(file_h_m_s) == 2:
-        file_h_m_s.insert(0, 0)
-    file_length_s = file_h_m_s[0] * 3600 + file_h_m_s[1] * 60 + file_h_m_s[2]
-    
-    if file_length_s > YT_LENGTH_LIMIT_S:
-        yt_length_limit_hms = time.strftime("%HH:%MM:%SS", time.gmtime(YT_LENGTH_LIMIT_S))
-        file_length_hms = time.strftime("%HH:%MM:%SS", time.gmtime(file_length_s))
-        raise gr.Error(f"Maximum YouTube length is {yt_length_limit_hms}, got {file_length_hms} YouTube video.")
-    
-    ydl_opts = {"outtmpl": filename, "format": "worstvideo[ext=mp4]+bestaudio[ext=m4a]/best[ext=mp4]/best"}
-    
-    with youtube_dl.YoutubeDL(ydl_opts) as ydl:
-        try:
-            ydl.download([yt_url])
-        except youtube_dl.utils.ExtractorError as err:
-            raise gr.Error(str(err))
-
-@spaces.GPU
-def yt_transcribe(yt_url, task, max_filesize=75.0):
-    html_embed_str = _return_yt_html_embed(yt_url)
-
-    with tempfile.TemporaryDirectory() as tmpdirname:
-        filepath = os.path.join(tmpdirname, "video.mp4")
-        download_yt_audio(yt_url, filepath)
-        with open(filepath, "rb") as f:
-            inputs = f.read()
-
-    inputs = ffmpeg_read(inputs, pipe.feature_extractor.sampling_rate)
-    inputs = {"array": inputs, "sampling_rate": pipe.feature_extractor.sampling_rate}
-
-    text = pipe(inputs, batch_size=BATCH_SIZE, generate_kwargs={"task": task}, return_timestamps=True)["text"]
-
-    return html_embed_str, text
-
+        raise gr.Error("No audio file submitted!")
+    return run_transcription(inputs, task)
 
+# ---------------------------------------------------------------------------
+# 1. 构建 Gradio 界面
+# ---------------------------------------------------------------------------
 demo = gr.Interface(
-    fn=transcribe,
+    fn=gradio_transcribe,
     inputs=[
         gr.Audio(type="filepath"),
         gr.Radio(["transcribe", "translate"], label="Task", value="transcribe"),
     ],
     outputs=gr.Textbox(lines=3),
-    title="Whisper Large V3: Transcribe Audio",
-    description=(
-        "Transcribe long-form microphone or audio inputs with the click of a button! Demo uses the OpenAI Whisper"
-        f" checkpoint [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) and 🤗 Transformers to transcribe audio files"
-        " of arbitrary length."
-    ),
+    title="Whisper Large V3 API Server",
     allow_flagging="never",
 )
 
-demo.launch()
+# ---------------------------------------------------------------------------
+# 2. 构建兼容 OpenAI 格式的 FastAPI 应用
+# ---------------------------------------------------------------------------
+app = FastAPI()
+
+async def handle_openai_audio_request(
+    file: UploadFile = File(...),
+    model: str = Form("whisper-1"),
+    response_format: str = Form("json"),
+    task: str = "transcribe",
+):
+    try:
+        # 读取上传的文件内容
+        audio_bytes = await file.read()
+        
+        # 调用推理核心
+        text = run_transcription(audio_bytes, task=task)
+
+        if response_format == "text":
+            return text
+        else:
+            # 默认返回标准的 OpenAI JSON 结构
+            return {"text": text}
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+# OpenAI 转录接口
+@app.post("/v1/audio/transcriptions")
+async def audio_transcriptions(
+    file: UploadFile = File(...),
+    model: str = Form("whisper-1"),
+    response_format: str = Form("json"),
+):
+    return await handle_openai_audio_request(file, model, response_format, task="transcribe")
+
+# OpenAI 翻译接口
+@app.post("/v1/audio/translations")
+async def audio_translations(
+    file: UploadFile = File(...),
+    model: str = Form("whisper-1"),
+    response_format: str = Form("json"),
+):
+    return await handle_openai_audio_request(file, model, response_format, task="translate")
+
+# ---------------------------------------------------------------------------
+# 3. 将 Gradio 挂载到 FastAPI 上
+# ---------------------------------------------------------------------------
+app = gr.mount_gradio_app(app, demo, path="/")