Update

README.md

@@ -4,7 +4,8 @@ emoji: 😻
 colorFrom: red
 colorTo: blue
 sdk: gradio
-sdk_version: 6.0.2
+sdk_version: 6.10.0
+python_version: "3.12.12"
 app_file: app.py
 pinned: false
 ---

app.py

@@ -12,6 +12,8 @@ from transformers import KyutaiSpeechToTextForConditionalGeneration, KyutaiSpeec
 device = "cuda" if torch.cuda.is_available() else "cpu"
 model_id = "kyutai/stt-2.6b-en-trfs"
 model = KyutaiSpeechToTextForConditionalGeneration.from_pretrained(model_id, device_map=device, torch_dtype="auto")
+model.generation_config.disable_compile = True
+model.generation_config.cache_implementation = "static"
 processor = KyutaiSpeechToTextProcessor.from_pretrained(model_id)
 
 SAMPLE_RATE = 24000
@@ -37,11 +39,10 @@ def transcribe(audio_path: str) -> str:
         data = data[:MAX_SAMPLE_SIZE]
         gr.Info(f"Audio file is too long. Truncating to {MAX_DURATION} seconds.")
 
-    inputs = processor(data)
+    inputs = processor(audio=data)
     inputs.to(device)
     output_tokens = model.generate(**inputs)
-    output = processor.batch_decode(output_tokens, skip_special_tokens=True)
-    return output[0]
+    return processor.decode(output_tokens, skip_special_tokens=True)[0]
 
 
 with gr.Blocks(fill_height=False) as demo:
@@ -61,7 +62,7 @@ with gr.Blocks(fill_height=False) as demo:
     """)  # noqa: RUF001
 
     # Main content
-    with gr.Group(elem_classes="main-card"):
+    with gr.Column(elem_classes="main-card"):
         # Audio input
         audio = gr.Audio(
             label="🎵 Audio Input",
@@ -88,26 +89,14 @@ with gr.Blocks(fill_height=False) as demo:
         )
 
     # Examples section
-    with gr.Group(elem_classes="examples-container"):
-        gr.Markdown("### 💡 Try These Examples")
-        gr.Examples(
-            examples=sorted(pathlib.Path("assets").glob("*.wav")) if pathlib.Path("assets").exists() else [],
-            inputs=audio,
-            outputs=output,
-            fn=transcribe,
-            examples_per_page=5,
-        )
-
-    # Footer
-    gr.HTML("""
-        <div class="footer-container">
-            <p>
-                Built with <a href="https://huggingface.co/spaces/akhaliq/anycoder" class="footer-link" target="_blank">anycoder</a> •
-                Powered by <a href="https://huggingface.co/kyutai/stt-2.6b-en-trfs" class="footer-link" target="_blank">Kyutai STT 2.6B</a>
-            </p>
-        </div>
-    """)
-
+    gr.Markdown("### 💡 Try These Examples", elem_classes="examples-container")
+    gr.Examples(
+        examples=sorted(pathlib.Path("assets").glob("*.wav")) if pathlib.Path("assets").exists() else [],
+        inputs=audio,
+        outputs=output,
+        fn=transcribe,
+        examples_per_page=5,
+    )
     # Event handlers
     transcribe_btn.click(
         fn=transcribe,

pyproject.toml

@@ -5,12 +5,20 @@ description = ""
 readme = "README.md"
 requires-python = ">=3.10"
 dependencies = [
-    "accelerate>=1.12.0",
-    "gradio[mcp]>=6.0.2",
+    "accelerate>=1.13.0",
+    "gradio[mcp]>=6.10.0",
     "librosa>=0.11.0",
-    "spaces>=0.44.0",
-    "torch==2.8.0",
-    "transformers==4.54.1",
+    "spaces>=0.48.1",
+    "torch==2.9.1",
+    "transformers==5.4.0",
+]
+
+[dependency-groups]
+dev = [
+    "ruff>=0.15.8",
+]
+hf-spaces = [
+    "datasets",
 ]
 
 [tool.ruff]

requirements.txt

@@ -1,29 +1,45 @@
 # This file was autogenerated by uv via the following command:
-#    uv pip compile pyproject.toml -o requirements.txt
-accelerate==1.12.0
-    # via kyutai-stt-2-6b-en (pyproject.toml)
+#    uv export --no-hashes --no-dev --group hf-spaces --no-emit-package typer-slim --no-emit-package spaces -o requirements.txt
+accelerate==1.13.0
+    # via kyutai-stt-2-6b-en
 aiofiles==24.1.0
     # via gradio
+aiohappyeyeballs==2.6.1
+    # via aiohttp
+aiohttp==3.13.4
+    # via fsspec
+aiosignal==1.4.0
+    # via aiohttp
 annotated-doc==0.0.4
-    # via fastapi
+    # via
+    #   fastapi
+    #   typer
 annotated-types==0.7.0
     # via pydantic
-anyio==4.12.0
+anyio==4.13.0
     # via
     #   gradio
     #   httpx
     #   mcp
     #   sse-starlette
     #   starlette
-attrs==25.4.0
+async-timeout==5.0.1 ; python_full_version < '3.11'
+    # via aiohttp
+attrs==26.1.0
     # via
+    #   aiohttp
     #   jsonschema
     #   referencing
+audioop-lts==0.2.2 ; python_full_version >= '3.13'
+    # via
+    #   gradio
+    #   standard-aifc
+    #   standard-sunau
 audioread==3.1.0
     # via librosa
 brotli==1.2.0
     # via gradio
-certifi==2025.11.12
+certifi==2026.2.25
     # via
     #   httpcore
     #   httpx
@@ -32,60 +48,81 @@ cffi==2.0.0
     # via
     #   cryptography
     #   soundfile
-charset-normalizer==3.4.4
+charset-normalizer==3.4.6
     # via requests
 click==8.3.1
     # via
     #   typer
     #   uvicorn
-cryptography==46.0.3
+colorama==0.4.6 ; sys_platform == 'win32'
+    # via
+    #   click
+    #   tqdm
+cryptography==46.0.6
     # via pyjwt
+datasets==4.8.4
 decorator==5.2.1
     # via librosa
-exceptiongroup==1.3.1
+dill==0.4.1
+    # via
+    #   datasets
+    #   multiprocess
+exceptiongroup==1.3.1 ; python_full_version < '3.11'
     # via anyio
-fastapi==0.123.7
+fastapi==0.135.2
     # via gradio
 ffmpy==1.0.0
     # via gradio
-filelock==3.20.0
+filelock==3.25.2
     # via
+    #   datasets
     #   huggingface-hub
     #   torch
-    #   transformers
-fsspec==2025.12.0
+frozenlist==1.8.0
     # via
+    #   aiohttp
+    #   aiosignal
+fsspec==2026.2.0
+    # via
+    #   datasets
     #   gradio-client
     #   huggingface-hub
     #   torch
-gradio==6.0.2
+gradio==6.10.0
     # via
-    #   kyutai-stt-2-6b-en (pyproject.toml)
+    #   kyutai-stt-2-6b-en
     #   spaces
-gradio-client==2.0.1
-    # via gradio
+gradio-client==2.4.0
+    # via
+    #   gradio
+    #   hf-gradio
 groovy==0.1.2
     # via gradio
 h11==0.16.0
     # via
     #   httpcore
     #   uvicorn
-hf-xet==1.2.0
+hf-gradio==0.3.0
+    # via gradio
+hf-xet==1.4.2 ; platform_machine == 'AMD64' or platform_machine == 'aarch64' or platform_machine == 'amd64' or platform_machine == 'arm64' or platform_machine == 'x86_64'
     # via huggingface-hub
 httpcore==1.0.9
     # via httpx
 httpx==0.28.1
     # via
+    #   datasets
     #   gradio
     #   gradio-client
+    #   huggingface-hub
     #   mcp
     #   safehttpx
     #   spaces
 httpx-sse==0.4.3
     # via mcp
-huggingface-hub==0.36.0
+huggingface-hub==1.8.0
     # via
     #   accelerate
+    #   datasets
     #   gradio
     #   gradio-client
     #   tokenizers
@@ -95,23 +132,24 @@ idna==3.11
     #   anyio
     #   httpx
     #   requests
+    #   yarl
 jinja2==3.1.6
     # via
     #   gradio
     #   torch
-joblib==1.5.2
+joblib==1.5.3
     # via
     #   librosa
     #   scikit-learn
-jsonschema==4.25.1
+jsonschema==4.26.0
     # via mcp
 jsonschema-specifications==2025.9.1
     # via jsonschema
-lazy-loader==0.4
+lazy-loader==0.5
     # via librosa
 librosa==0.11.0
-    # via kyutai-stt-2-6b-en (pyproject.toml)
-llvmlite==0.45.1
+    # via kyutai-stt-2-6b-en
+llvmlite==0.46.0
     # via numba
 markdown-it-py==4.0.0
     # via rich
@@ -119,7 +157,7 @@ markupsafe==3.0.3
     # via
     #   gradio
     #   jinja2
-mcp==1.23.1
+mcp==1.26.0
     # via gradio
 mdurl==0.1.2
     # via markdown-it-py
@@ -127,13 +165,35 @@ mpmath==1.3.0
     # via sympy
 msgpack==1.1.2
     # via librosa
-networkx==3.4.2
+multidict==6.7.1
+    # via
+    #   aiohttp
+    #   yarl
+multiprocess==0.70.19
+    # via datasets
+networkx==3.4.2 ; python_full_version < '3.11'
     # via torch
-numba==0.62.1
+networkx==3.6.1 ; python_full_version >= '3.11'
+    # via torch
+numba==0.64.0
     # via librosa
-numpy==2.2.6
+numpy==2.2.6 ; python_full_version < '3.11'
+    # via
+    #   accelerate
+    #   datasets
+    #   gradio
+    #   librosa
+    #   numba
+    #   pandas
+    #   scikit-learn
+    #   scipy
+    #   soundfile
+    #   soxr
+    #   transformers
+numpy==2.4.4 ; python_full_version >= '3.11'
     # via
     #   accelerate
+    #   datasets
     #   gradio
     #   librosa
     #   numba
@@ -143,48 +203,51 @@ numpy==2.2.6
     #   soundfile
     #   soxr
     #   transformers
-nvidia-cublas-cu12==12.8.4.1
+nvidia-cublas-cu12==12.8.4.1 ; platform_machine == 'x86_64' and sys_platform == 'linux'
     # via
     #   nvidia-cudnn-cu12
     #   nvidia-cusolver-cu12
     #   torch
-nvidia-cuda-cupti-cu12==12.8.90
+nvidia-cuda-cupti-cu12==12.8.90 ; platform_machine == 'x86_64' and sys_platform == 'linux'
     # via torch
-nvidia-cuda-nvrtc-cu12==12.8.93
+nvidia-cuda-nvrtc-cu12==12.8.93 ; platform_machine == 'x86_64' and sys_platform == 'linux'
     # via torch
-nvidia-cuda-runtime-cu12==12.8.90
+nvidia-cuda-runtime-cu12==12.8.90 ; platform_machine == 'x86_64' and sys_platform == 'linux'
     # via torch
-nvidia-cudnn-cu12==9.10.2.21
+nvidia-cudnn-cu12==9.10.2.21 ; platform_machine == 'x86_64' and sys_platform == 'linux'
     # via torch
-nvidia-cufft-cu12==11.3.3.83
+nvidia-cufft-cu12==11.3.3.83 ; platform_machine == 'x86_64' and sys_platform == 'linux'
     # via torch
-nvidia-cufile-cu12==1.13.1.3
+nvidia-cufile-cu12==1.13.1.3 ; platform_machine == 'x86_64' and sys_platform == 'linux'
     # via torch
-nvidia-curand-cu12==10.3.9.90
+nvidia-curand-cu12==10.3.9.90 ; platform_machine == 'x86_64' and sys_platform == 'linux'
     # via torch
-nvidia-cusolver-cu12==11.7.3.90
+nvidia-cusolver-cu12==11.7.3.90 ; platform_machine == 'x86_64' and sys_platform == 'linux'
     # via torch
-nvidia-cusparse-cu12==12.5.8.93
+nvidia-cusparse-cu12==12.5.8.93 ; platform_machine == 'x86_64' and sys_platform == 'linux'
     # via
     #   nvidia-cusolver-cu12
     #   torch
-nvidia-cusparselt-cu12==0.7.1
+nvidia-cusparselt-cu12==0.7.1 ; platform_machine == 'x86_64' and sys_platform == 'linux'
     # via torch
-nvidia-nccl-cu12==2.27.3
+nvidia-nccl-cu12==2.27.5 ; platform_machine == 'x86_64' and sys_platform == 'linux'
     # via torch
-nvidia-nvjitlink-cu12==12.8.93
+nvidia-nvjitlink-cu12==12.8.93 ; platform_machine == 'x86_64' and sys_platform == 'linux'
     # via
     #   nvidia-cufft-cu12
     #   nvidia-cusolver-cu12
     #   nvidia-cusparse-cu12
     #   torch
-nvidia-nvtx-cu12==12.8.90
+nvidia-nvshmem-cu12==3.3.20 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+    # via torch
+nvidia-nvtx-cu12==12.8.90 ; platform_machine == 'x86_64' and sys_platform == 'linux'
     # via torch
-orjson==3.11.4
+orjson==3.11.7
     # via gradio
-packaging==25.0
+packaging==26.0
     # via
     #   accelerate
+    #   datasets
     #   gradio
     #   gradio-client
     #   huggingface-hub
@@ -192,21 +255,33 @@ packaging==25.0
     #   pooch
     #   spaces
     #   transformers
-pandas==2.3.3
-    # via gradio
-pillow==12.0.0
+pandas==2.3.3 ; python_full_version < '3.11'
+    # via
+    #   datasets
+    #   gradio
+pandas==3.0.2 ; python_full_version >= '3.11'
+    # via
+    #   datasets
+    #   gradio
+pillow==12.1.1
     # via gradio
-platformdirs==4.5.0
+platformdirs==4.9.4
     # via pooch
-pooch==1.8.2
+pooch==1.9.0
     # via librosa
+propcache==0.4.1
+    # via
+    #   aiohttp
+    #   yarl
 psutil==5.9.8
     # via
     #   accelerate
     #   spaces
-pycparser==2.23
+pyarrow==23.0.1
+    # via datasets
+pycparser==3.0 ; implementation_name != 'PyPy'
     # via cffi
-pydantic==2.12.4
+pydantic==2.12.5
     # via
     #   fastapi
     #   gradio
@@ -215,27 +290,32 @@ pydantic==2.12.4
     #   spaces
 pydantic-core==2.41.5
     # via pydantic
-pydantic-settings==2.12.0
+pydantic-settings==2.13.1
     # via mcp
 pydub==0.25.1
     # via gradio
-pygments==2.19.2
+pygments==2.20.0
     # via rich
-pyjwt==2.10.1
+pyjwt==2.12.1
     # via mcp
 python-dateutil==2.9.0.post0
     # via pandas
-python-dotenv==1.2.1
+python-dotenv==1.2.2
     # via pydantic-settings
-python-multipart==0.0.20
+python-multipart==0.0.22
     # via
     #   gradio
     #   mcp
-pytz==2025.2
-    # via pandas
+pytz==2026.1.post1
+    # via
+    #   gradio
+    #   pandas
+pywin32==311 ; sys_platform == 'win32'
+    # via mcp
 pyyaml==6.0.3
     # via
     #   accelerate
+    #   datasets
     #   gradio
     #   huggingface-hub
     #   transformers
@@ -243,15 +323,14 @@ referencing==0.37.0
     # via
     #   jsonschema
     #   jsonschema-specifications
-regex==2025.11.3
+regex==2026.3.32
     # via transformers
-requests==2.32.5
+requests==2.33.1
     # via
-    #   huggingface-hub
+    #   datasets
     #   pooch
     #   spaces
-    #   transformers
-rich==14.2.0
+rich==14.3.3
     # via typer
 rpds-py==0.30.0
     # via
@@ -263,16 +342,22 @@ safetensors==0.7.0
     # via
     #   accelerate
     #   transformers
-scikit-learn==1.7.2
+scikit-learn==1.7.2 ; python_full_version < '3.11'
     # via librosa
-scipy==1.15.3
+scikit-learn==1.8.0 ; python_full_version >= '3.11'
+    # via librosa
+scipy==1.15.3 ; python_full_version < '3.11'
+    # via
+    #   librosa
+    #   scikit-learn
+scipy==1.17.1 ; python_full_version >= '3.11'
     # via
     #   librosa
     #   scikit-learn
 semantic-version==2.10.0
     # via gradio
-setuptools==80.9.0
-    # via triton
+setuptools==82.0.1 ; python_full_version >= '3.12'
+    # via torch
 shellingham==1.5.4
     # via typer
 six==1.17.0
@@ -281,39 +366,54 @@ soundfile==0.13.1
     # via librosa
 soxr==1.0.0
     # via librosa
-spaces==0.44.0
-    # via kyutai-stt-2-6b-en (pyproject.toml)
-sse-starlette==3.0.3
+sse-starlette==3.3.4
     # via mcp
-starlette==0.50.0
+standard-aifc==3.13.0 ; python_full_version >= '3.13'
+    # via
+    #   audioread
+    #   librosa
+standard-chunk==3.13.0 ; python_full_version >= '3.13'
+    # via standard-aifc
+standard-sunau==3.13.0 ; python_full_version >= '3.13'
+    # via
+    #   audioread
+    #   librosa
+starlette==0.52.1
     # via
     #   fastapi
     #   gradio
     #   mcp
+    #   sse-starlette
 sympy==1.14.0
     # via torch
 threadpoolctl==3.6.0
     # via scikit-learn
-tokenizers==0.21.4
+tokenizers==0.22.2
     # via transformers
 tomlkit==0.13.3
     # via gradio
-torch==2.8.0
+torch==2.9.1
     # via
-    #   kyutai-stt-2-6b-en (pyproject.toml)
     #   accelerate
-tqdm==4.67.1
+    #   kyutai-stt-2-6b-en
+tqdm==4.67.3
     # via
+    #   datasets
     #   huggingface-hub
     #   transformers
-transformers==4.54.1
-    # via kyutai-stt-2-6b-en (pyproject.toml)
-triton==3.4.0
+transformers==5.4.0
+    # via kyutai-stt-2-6b-en
+triton==3.5.1 ; platform_machine == 'x86_64' and sys_platform == 'linux'
     # via torch
-typer==0.20.0
-    # via gradio
+typer==0.24.1
+    # via
+    #   gradio
+    #   hf-gradio
+    #   huggingface-hub
+    #   transformers
 typing-extensions==4.15.0
     # via
+    #   aiosignal
     #   anyio
     #   cryptography
     #   exceptiongroup
@@ -323,25 +423,31 @@ typing-extensions==4.15.0
     #   huggingface-hub
     #   librosa
     #   mcp
+    #   multidict
     #   pydantic
     #   pydantic-core
+    #   pyjwt
     #   referencing
     #   spaces
     #   starlette
     #   torch
-    #   typer
     #   typing-inspection
     #   uvicorn
 typing-inspection==0.4.2
     # via
+    #   fastapi
     #   mcp
     #   pydantic
     #   pydantic-settings
-tzdata==2025.2
+tzdata==2025.3 ; python_full_version < '3.11' or sys_platform == 'emscripten' or sys_platform == 'win32'
     # via pandas
-urllib3==2.5.0
+urllib3==2.6.3
     # via requests
-uvicorn==0.38.0
+uvicorn==0.42.0
     # via
     #   gradio
     #   mcp
+xxhash==3.6.0
+    # via datasets
+yarl==1.23.0
+    # via aiohttp

style.css

@@ -28,12 +28,21 @@
     font-weight: 400;
 }
 
-/* Main content card */
+/* Info banner */
+.info-banner {
+    border-left: 4px solid #3b82f6;
+    padding: 1rem;
+    border-radius: 0.5rem;
+    margin-bottom: 1.5rem;
+    font-size: 0.875rem;
+    background: var(--background-fill-secondary);
+    color: var(--body-text-color);
+}
+
+/* Main content card — let gr.Group handle background */
 .main-card {
-    background: white;
     border-radius: 1rem;
     padding: 1.5rem;
-    box-shadow: 0 4px 6px -1px rgba(0, 0, 0, 0.1), 0 2px 4px -1px rgba(0, 0, 0, 0.06);
     margin-bottom: 1.5rem;
 }
 
@@ -42,10 +51,9 @@
     margin-bottom: 1.5rem;
 }
 
-/* Transcription output styling */
+/* Transcription output styling — no hardcoded background */
 .transcription-output {
     min-height: 120px;
-    background: #f8fafc;
     border-radius: 0.75rem;
     padding: 1rem;
     font-size: 1rem;
@@ -57,27 +65,6 @@
     margin-top: 2rem;
 }
 
-/* Footer */
-.footer-container {
-    text-align: center;
-    padding: 1.5rem 1rem;
-    margin-top: 2rem;
-    border-top: 1px solid #e2e8f0;
-    font-size: 0.875rem;
-    color: #64748b;
-}
-
-.footer-link {
-    color: #667eea;
-    text-decoration: none;
-    font-weight: 500;
-    transition: color 0.2s;
-}
-
-.footer-link:hover {
-    color: #764ba2;
-}
-
 /* Mobile optimizations */
 @media (max-width: 640px) {
     .header-title {
@@ -111,14 +98,3 @@
     transform: translateY(-2px) !important;
     box-shadow: 0 10px 20px rgba(102, 126, 234, 0.3) !important;
 }
-
-/* Info banner */
-.info-banner {
-    background: #eff6ff;
-    border-left: 4px solid #3b82f6;
-    padding: 1rem;
-    border-radius: 0.5rem;
-    margin-bottom: 1.5rem;
-    font-size: 0.875rem;
-    color: #1e40af;
-}