Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.
Python asraudiochineseemotion-recognitionfunasrmcp-servermultilingual-asropenai-compatible-apiparaformerpunctuationpytorchreal-time-asrspeaker-diarizationspeech-recognitionspeech-to-textstreaming-asrtranscriptionvllmvoice-activity-detectionwhisper-alternative
feat: auto-select model based on device (GPU=fun-asr-nano, CPU=sensevoice)
游雁b61d9f5b4978
diff --git a/funasr/bin/_server_app.py b/funasr/bin/_server_app.pyindex dde92f0bf..0e1357d3e 100644--- a/funasr/bin/_server_app.py+++ b/funasr/bin/_server_app.py@@ -37,7 +37,9 @@ MODEL_CONFIGS = {}-def create_app(device: str = "cuda", preload_model: str = "sensevoice") -> FastAPI:+def create_app(device: str = "cuda", preload_model: str = "auto") -> FastAPI:+ if preload_model == "auto":+ preload_model = "fun-asr-nano" if device.startswith("cuda") else "sensevoice"app = FastAPI(title="FunASR Server", version="1.3.2")app.state.models = {}app.state.device = devicediff --git a/funasr/bin/server.py b/funasr/bin/server.pyindex 7a4e6dd86..57763996e 100644--- a/funasr/bin/server.py+++ b/funasr/bin/server.py@@ -31,7 +31,7 @@ Then use with OpenAI SDK:parser.add_argument("--host", default="0.0.0.0", help="Bind address (default: 0.0.0.0)")parser.add_argument("--port", type=int, default=8000, help="Port (default: 8000)")parser.add_argument("--device", default="cuda", help="Device: cuda, cpu, mps (default: cuda)")- parser.add_argument("--model", default="sensevoice", help="Pre-load model: sensevoice, paraformer, fun-asr-nano (default: sensevoice)")+ parser.add_argument("--model", default="auto", help="Pre-load model: auto (GPU=fun-asr-nano, CPU=sensevoice), sensevoice, paraformer, fun-asr-nano")args = parser.parse_args()try: