feat(nemo): add Nemo (only asr for now) backend

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
2026-02-06 20:54:37 -05:00 · 2026-02-07 00:00:20 +01:00
parent ae2689936a
commit ec7e5129e7
20 changed files with 534 additions and 2 deletions
--- a/.github/workflows/backend.yml
+++ b/.github/workflows/backend.yml
@@ -184,6 +184,19 @@ jobs:
            dockerfile: "./backend/Dockerfile.python"
            context: "./"
            ubuntu-version: '2404'
+          - build-type: 'cublas'
+            cuda-major-version: "12"
+            cuda-minor-version: "8"
+            platforms: 'linux/amd64'
+            tag-latest: 'auto'
+            tag-suffix: '-gpu-nvidia-cuda-12-nemo'
+            runs-on: 'ubuntu-latest'
+            base-image: "ubuntu:24.04"
+            skip-drivers: 'false'
+            backend: "nemo"
+            dockerfile: "./backend/Dockerfile.python"
+            context: "./"
+            ubuntu-version: '2404'
          - build-type: 'cublas'
            cuda-major-version: "12"
            cuda-minor-version: "8"
@@ -536,6 +549,19 @@ jobs:
            dockerfile: "./backend/Dockerfile.python"
            context: "./"
            ubuntu-version: '2404'
+          - build-type: 'cublas'
+            cuda-major-version: "13"
+            cuda-minor-version: "0"
+            platforms: 'linux/amd64'
+            tag-latest: 'auto'
+            tag-suffix: '-gpu-nvidia-cuda-13-nemo'
+            runs-on: 'ubuntu-latest'
+            base-image: "ubuntu:24.04"
+            skip-drivers: 'false'
+            backend: "nemo"
+            dockerfile: "./backend/Dockerfile.python"
+            context: "./"
+            ubuntu-version: '2404'
          - build-type: 'cublas'
            cuda-major-version: "13"
            cuda-minor-version: "0"
@@ -1045,6 +1071,19 @@ jobs:
            dockerfile: "./backend/Dockerfile.python"
            context: "./"
            ubuntu-version: '2404'
+          - build-type: 'hipblas'
+            cuda-major-version: ""
+            cuda-minor-version: ""
+            platforms: 'linux/amd64'
+            tag-latest: 'auto'
+            tag-suffix: '-gpu-rocm-hipblas-nemo'
+            runs-on: 'arc-runner-set'
+            base-image: "rocm/dev-ubuntu-24.04:6.4.4"
+            skip-drivers: 'false'
+            backend: "nemo"
+            dockerfile: "./backend/Dockerfile.python"
+            context: "./"
+            ubuntu-version: '2404'
          - build-type: 'hipblas'
            cuda-major-version: ""
            cuda-minor-version: ""
@@ -1372,6 +1411,19 @@ jobs:
            dockerfile: "./backend/Dockerfile.python"
            context: "./"
            ubuntu-version: '2404'
+          - build-type: 'intel'
+            cuda-major-version: ""
+            cuda-minor-version: ""
+            platforms: 'linux/amd64'
+            tag-latest: 'auto'
+            tag-suffix: '-gpu-intel-nemo'
+            runs-on: 'arc-runner-set'
+            base-image: "intel/oneapi-basekit:2025.3.0-0-devel-ubuntu24.04"
+            skip-drivers: 'false'
+            backend: "nemo"
+            dockerfile: "./backend/Dockerfile.python"
+            context: "./"
+            ubuntu-version: '2404'
          - build-type: 'intel'
            cuda-major-version: ""
            cuda-minor-version: ""
@@ -1797,6 +1849,19 @@ jobs:
            dockerfile: "./backend/Dockerfile.python"
            context: "./"
            ubuntu-version: '2404'
+          - build-type: ''
+            cuda-major-version: ""
+            cuda-minor-version: ""
+            platforms: 'linux/amd64,linux/arm64'
+            tag-latest: 'auto'
+            tag-suffix: '-cpu-nemo'
+            runs-on: 'ubuntu-latest'
+            base-image: "ubuntu:24.04"
+            skip-drivers: 'false'
+            backend: "nemo"
+            dockerfile: "./backend/Dockerfile.python"
+            context: "./"
+            ubuntu-version: '2404'
          - build-type: ''
            cuda-major-version: ""
            cuda-minor-version: ""
@@ -1886,6 +1951,9 @@ jobs:
          - backend: "qwen-asr"
            tag-suffix: "-metal-darwin-arm64-qwen-asr"
            build-type: "mps"
+          - backend: "nemo"
+            tag-suffix: "-metal-darwin-arm64-nemo"
+            build-type: "mps"
          - backend: "qwen-tts"
            tag-suffix: "-metal-darwin-arm64-qwen-tts"
            build-type: "mps"
--- a/.github/workflows/test-extra.yml
+++ b/.github/workflows/test-extra.yml
@@ -323,6 +323,25 @@ jobs:
        run: |
          make --jobs=5 --output-sync=target -C backend/python/qwen-asr
          make --jobs=5 --output-sync=target -C backend/python/qwen-asr test
+  tests-nemo:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Clone
+        uses: actions/checkout@v6
+        with:
+          submodules: true
+      - name: Dependencies
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y build-essential ffmpeg sox
+          sudo apt-get install -y ca-certificates cmake curl patch python3-pip
+          # Install UV
+          curl -LsSf https://astral.sh/uv/install.sh | sh
+          pip install --user --no-cache-dir grpcio-tools==1.64.1
+      - name: Test nemo
+        run: |
+          make --jobs=5 --output-sync=target -C backend/python/nemo
+          make --jobs=5 --output-sync=target -C backend/python/nemo test
  tests-voxcpm:
    runs-on: ubuntu-latest
    steps:
--- a/8
+++ b/8
@@ -1,5 +1,5 @@
 # Disable parallel execution for backend builds
-.NOTPARALLEL: backends/diffusers backends/llama-cpp backends/outetts backends/piper backends/stablediffusion-ggml backends/whisper backends/faster-whisper backends/silero-vad backends/local-store backends/huggingface backends/rfdetr backends/kitten-tts backends/kokoro backends/chatterbox backends/llama-cpp-darwin backends/neutts build-darwin-python-backend build-darwin-go-backend backends/mlx backends/diffuser-darwin backends/mlx-vlm backends/mlx-audio backends/stablediffusion-ggml-darwin backends/vllm backends/vllm-omni backends/moonshine backends/pocket-tts backends/qwen-tts backends/qwen-asr backends/voxcpm backends/whisperx backends/ace-step
+.NOTPARALLEL: backends/diffusers backends/llama-cpp backends/outetts backends/piper backends/stablediffusion-ggml backends/whisper backends/faster-whisper backends/silero-vad backends/local-store backends/huggingface backends/rfdetr backends/kitten-tts backends/kokoro backends/chatterbox backends/llama-cpp-darwin backends/neutts build-darwin-python-backend build-darwin-go-backend backends/mlx backends/diffuser-darwin backends/mlx-vlm backends/mlx-audio backends/stablediffusion-ggml-darwin backends/vllm backends/vllm-omni backends/moonshine backends/pocket-tts backends/qwen-tts backends/qwen-asr backends/nemo backends/voxcpm backends/whisperx backends/ace-step

 GOCMD=go
 GOTEST=$(GOCMD) test
@@ -318,6 +318,7 @@ prepare-test-extra: protogen-python
 	$(MAKE) -C backend/python/pocket-tts
 	$(MAKE) -C backend/python/qwen-tts
 	$(MAKE) -C backend/python/qwen-asr
+	$(MAKE) -C backend/python/nemo
 	$(MAKE) -C backend/python/voxcpm
 	$(MAKE) -C backend/python/whisperx
 	$(MAKE) -C backend/python/ace-step
@@ -334,6 +335,7 @@ test-extra: prepare-test-extra
 	$(MAKE) -C backend/python/pocket-tts test
 	$(MAKE) -C backend/python/qwen-tts test
 	$(MAKE) -C backend/python/qwen-asr test
+	$(MAKE) -C backend/python/nemo test
 	$(MAKE) -C backend/python/voxcpm test
 	$(MAKE) -C backend/python/whisperx test
 	$(MAKE) -C backend/python/ace-step test
@@ -471,6 +473,7 @@ BACKEND_MOONSHINE = moonshine|python|.|false|true
 BACKEND_POCKET_TTS = pocket-tts|python|.|false|true
 BACKEND_QWEN_TTS = qwen-tts|python|.|false|true
 BACKEND_QWEN_ASR = qwen-asr|python|.|false|true
+BACKEND_NEMO = nemo|python|.|false|true
 BACKEND_VOXCPM = voxcpm|python|.|false|true
 BACKEND_WHISPERX = whisperx|python|.|false|true
 BACKEND_ACE_STEP = ace-step|python|.|false|true
@@ -521,6 +524,7 @@ $(eval $(call generate-docker-build-target,$(BACKEND_MOONSHINE)))
 $(eval $(call generate-docker-build-target,$(BACKEND_POCKET_TTS)))
 $(eval $(call generate-docker-build-target,$(BACKEND_QWEN_TTS)))
 $(eval $(call generate-docker-build-target,$(BACKEND_QWEN_ASR)))
+$(eval $(call generate-docker-build-target,$(BACKEND_NEMO)))
 $(eval $(call generate-docker-build-target,$(BACKEND_VOXCPM)))
 $(eval $(call generate-docker-build-target,$(BACKEND_WHISPERX)))
 $(eval $(call generate-docker-build-target,$(BACKEND_ACE_STEP)))
@@ -529,7 +533,7 @@ $(eval $(call generate-docker-build-target,$(BACKEND_ACE_STEP)))
 docker-save-%: backend-images
 	docker save local-ai-backend:$* -o backend-images/$*.tar

-docker-build-backends: docker-build-llama-cpp docker-build-rerankers docker-build-vllm docker-build-vllm-omni docker-build-transformers docker-build-outetts docker-build-diffusers docker-build-kokoro docker-build-faster-whisper docker-build-coqui docker-build-chatterbox docker-build-vibevoice docker-build-moonshine docker-build-pocket-tts docker-build-qwen-tts docker-build-qwen-asr docker-build-voxcpm docker-build-whisperx docker-build-ace-step
+docker-build-backends: docker-build-llama-cpp docker-build-rerankers docker-build-vllm docker-build-vllm-omni docker-build-transformers docker-build-outetts docker-build-diffusers docker-build-kokoro docker-build-faster-whisper docker-build-coqui docker-build-chatterbox docker-build-vibevoice docker-build-moonshine docker-build-pocket-tts docker-build-qwen-tts docker-build-qwen-asr docker-build-nemo docker-build-voxcpm docker-build-whisperx docker-build-ace-step

 ########################################################
 ### Mock Backend for E2E Tests
--- a/backend/index.yaml
+++ b/backend/index.yaml
@@ -536,6 +536,27 @@
    nvidia-l4t-cuda-12: "nvidia-l4t-qwen-asr"
    nvidia-l4t-cuda-13: "cuda13-nvidia-l4t-arm64-qwen-asr"
  icon: https://cdn-avatars.huggingface.co/v1/production/uploads/620760a26e3b7210c2ff1943/-s1gyJfvbE1RgO5iBeNOi.png
+- &nemo
+  urls:
+    - https://github.com/NVIDIA/NeMo
+  description: |
+    NVIDIA NEMO Toolkit for ASR provides state-of-the-art automatic speech recognition models including Parakeet models for various languages and use cases.
+  tags:
+    - speech-recognition
+    - ASR
+    - NVIDIA
+  license: apache-2.0
+  name: "nemo"
+  alias: "nemo"
+  capabilities:
+    nvidia: "cuda12-nemo"
+    intel: "intel-nemo"
+    amd: "rocm-nemo"
+    metal: "metal-nemo"
+    default: "cpu-nemo"
+    nvidia-cuda-13: "cuda13-nemo"
+    nvidia-cuda-12: "cuda12-nemo"
+  icon: https://www.nvidia.com/favicon.ico
 - &voxcpm
  urls:
    - https://github.com/ModelBest/VoxCPM
@@ -2337,6 +2358,77 @@
  uri: "quay.io/go-skynet/local-ai-backends:master-metal-darwin-arm64-qwen-asr"
  mirrors:
    - localai/localai-backends:master-metal-darwin-arm64-qwen-asr
+## nemo
+- !!merge <<: *nemo
+  name: "nemo-development"
+  capabilities:
+    nvidia: "cuda12-nemo-development"
+    intel: "intel-nemo-development"
+    amd: "rocm-nemo-development"
+    metal: "metal-nemo-development"
+    default: "cpu-nemo-development"
+    nvidia-cuda-13: "cuda13-nemo-development"
+    nvidia-cuda-12: "cuda12-nemo-development"
+- !!merge <<: *nemo
+  name: "cpu-nemo"
+  uri: "quay.io/go-skynet/local-ai-backends:latest-cpu-nemo"
+  mirrors:
+    - localai/localai-backends:latest-cpu-nemo
+- !!merge <<: *nemo
+  name: "cpu-nemo-development"
+  uri: "quay.io/go-skynet/local-ai-backends:master-cpu-nemo"
+  mirrors:
+    - localai/localai-backends:master-cpu-nemo
+- !!merge <<: *nemo
+  name: "cuda12-nemo"
+  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-nvidia-cuda-12-nemo"
+  mirrors:
+    - localai/localai-backends:latest-gpu-nvidia-cuda-12-nemo
+- !!merge <<: *nemo
+  name: "cuda12-nemo-development"
+  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-12-nemo"
+  mirrors:
+    - localai/localai-backends:master-gpu-nvidia-cuda-12-nemo
+- !!merge <<: *nemo
+  name: "cuda13-nemo"
+  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-nvidia-cuda-13-nemo"
+  mirrors:
+    - localai/localai-backends:latest-gpu-nvidia-cuda-13-nemo
+- !!merge <<: *nemo
+  name: "cuda13-nemo-development"
+  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-nvidia-cuda-13-nemo"
+  mirrors:
+    - localai/localai-backends:master-gpu-nvidia-cuda-13-nemo
+- !!merge <<: *nemo
+  name: "intel-nemo"
+  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-intel-nemo"
+  mirrors:
+    - localai/localai-backends:latest-gpu-intel-nemo
+- !!merge <<: *nemo
+  name: "intel-nemo-development"
+  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-intel-nemo"
+  mirrors:
+    - localai/localai-backends:master-gpu-intel-nemo
+- !!merge <<: *nemo
+  name: "rocm-nemo"
+  uri: "quay.io/go-skynet/local-ai-backends:latest-gpu-rocm-hipblas-nemo"
+  mirrors:
+    - localai/localai-backends:latest-gpu-rocm-hipblas-nemo
+- !!merge <<: *nemo
+  name: "rocm-nemo-development"
+  uri: "quay.io/go-skynet/local-ai-backends:master-gpu-rocm-hipblas-nemo"
+  mirrors:
+    - localai/localai-backends:master-gpu-rocm-hipblas-nemo
+- !!merge <<: *nemo
+  name: "metal-nemo"
+  uri: "quay.io/go-skynet/local-ai-backends:latest-metal-darwin-arm64-nemo"
+  mirrors:
+    - localai/localai-backends:latest-metal-darwin-arm64-nemo
+- !!merge <<: *nemo
+  name: "metal-nemo-development"
+  uri: "quay.io/go-skynet/local-ai-backends:master-metal-darwin-arm64-nemo"
+  mirrors:
+    - localai/localai-backends:master-metal-darwin-arm64-nemo
 ## voxcpm
 - !!merge <<: *voxcpm
  name: "voxcpm-development"
--- a/backend/python/nemo/Makefile
+++ b/backend/python/nemo/Makefile
@@ -0,0 +1,23 @@
+.PHONY: nemo-asr
+nemo-asr:
+	bash install.sh
+
+.PHONY: run
+run: nemo-asr
+	@echo "Running nemo-asr..."
+	bash run.sh
+	@echo "nemo-asr run."
+
+.PHONY: test
+test: nemo-asr
+	@echo "Testing nemo-asr..."
+	bash test.sh
+	@echo "nemo-asr tested."
+
+.PHONY: protogen-clean
+protogen-clean:
+	$(RM) backend_pb2_grpc.py backend_pb2.py
+
+.PHONY: clean
+clean: protogen-clean
+	rm -rf venv __pycache__
--- a/backend/python/nemo/backend.py
+++ b/backend/python/nemo/backend.py
@@ -0,0 +1,147 @@
+#!/usr/bin/env python3
+"""
+gRPC server of LocalAI for NVIDIA NEMO Toolkit ASR.
+"""
+from concurrent import futures
+import time
+import argparse
+import signal
+import sys
+import os
+import backend_pb2
+import backend_pb2_grpc
+import torch
+import nemo.collections.asr as nemo_asr
+
+import grpc
+
+
+def is_float(s):
+    try:
+        float(s)
+        return True
+    except ValueError:
+        return False
+
+
+def is_int(s):
+    try:
+        int(s)
+        return True
+    except ValueError:
+        return False
+
+
+_ONE_DAY_IN_SECONDS = 60 * 60 * 24
+MAX_WORKERS = int(os.environ.get('PYTHON_GRPC_MAX_WORKERS', '1'))
+
+
+class BackendServicer(backend_pb2_grpc.BackendServicer):
+    def Health(self, request, context):
+        return backend_pb2.Reply(message=bytes("OK", 'utf-8'))
+
+    def LoadModel(self, request, context):
+        if torch.cuda.is_available():
+            device = "cuda"
+        else:
+            device = "cpu"
+        mps_available = hasattr(torch.backends, "mps") and torch.backends.mps.is_available()
+        if mps_available:
+            device = "mps"
+        if not torch.cuda.is_available() and request.CUDA:
+            return backend_pb2.Result(success=False, message="CUDA is not available")
+
+        self.device = device
+        self.options = {}
+
+        for opt in request.Options:
+            if ":" not in opt:
+                continue
+            key, value = opt.split(":", 1)
+            if is_float(value):
+                value = float(value)
+            elif is_int(value):
+                value = int(value)
+            elif value.lower() in ["true", "false"]:
+                value = value.lower() == "true"
+            self.options[key] = value
+
+        model_name = request.Model or "nvidia/parakeet-tdt-0.6b-v3"
+
+        try:
+            print(f"Loading NEMO ASR model from {model_name}", file=sys.stderr)
+            self.model = nemo_asr.models.ASRModel.from_pretrained(model_name=model_name)
+            print("NEMO ASR model loaded successfully", file=sys.stderr)
+        except Exception as err:
+            print(f"[ERROR] LoadModel failed: {err}", file=sys.stderr)
+            import traceback
+            traceback.print_exc(file=sys.stderr)
+            return backend_pb2.Result(success=False, message=str(err))
+
+        return backend_pb2.Result(message="Model loaded successfully", success=True)
+
+    def AudioTranscription(self, request, context):
+        result_segments = []
+        text = ""
+        try:
+            audio_path = request.dst
+            if not audio_path or not os.path.exists(audio_path):
+                print(f"Error: Audio file not found: {audio_path}", file=sys.stderr)
+                return backend_pb2.TranscriptResult(segments=[], text="")
+
+            # NEMO's transcribe method accepts a list of audio paths and returns a list of transcripts
+            results = self.model.transcribe([audio_path])
+
+            if not results or len(results) == 0:
+                return backend_pb2.TranscriptResult(segments=[], text="")
+
+            # Get the transcript text from the first result
+            text = results[0]
+            if text:
+                # Create a single segment with the full transcription
+                result_segments.append(backend_pb2.TranscriptSegment(
+                    id=0, start=0, end=0, text=text
+                ))
+
+        except Exception as err:
+            print(f"Error in AudioTranscription: {err}", file=sys.stderr)
+            import traceback
+            traceback.print_exc(file=sys.stderr)
+            return backend_pb2.TranscriptResult(segments=[], text="")
+
+        return backend_pb2.TranscriptResult(segments=result_segments, text=text)
+
+
+def serve(address):
+    server = grpc.server(
+        futures.ThreadPoolExecutor(max_workers=MAX_WORKERS),
+        options=[
+            ('grpc.max_message_length', 50 * 1024 * 1024),
+            ('grpc.max_send_message_length', 50 * 1024 * 1024),
+            ('grpc.max_receive_message_length', 50 * 1024 * 1024),
+        ])
+    backend_pb2_grpc.add_BackendServicer_to_server(BackendServicer(), server)
+    server.add_insecure_port(address)
+    server.start()
+    print("Server started. Listening on: " + address, file=sys.stderr)
+
+    def signal_handler(sig, frame):
+        print("Received termination signal. Shutting down...")
+        server.stop(0)
+        sys.exit(0)
+
+    signal.signal(signal.SIGINT, signal_handler)
+    signal.signal(signal.SIGTERM, signal_handler)
+
+    try:
+        while True:
+            time.sleep(_ONE_DAY_IN_SECONDS)
+    except KeyboardInterrupt:
+        server.stop(0)
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Run the gRPC server.")
+    parser.add_argument("--addr", default="localhost:50051", help="The address to bind the server to.")
+    args = parser.parse_args()
+    serve(args.addr)
--- a/backend/python/nemo/install.sh
+++ b/backend/python/nemo/install.sh
@@ -0,0 +1,21 @@
+#!/bin/bash
+set -e
+
+EXTRA_PIP_INSTALL_FLAGS="--no-build-isolation"
+
+backend_dir=$(dirname $0)
+if [ -d $backend_dir/common ]; then
+    source $backend_dir/common/libbackend.sh
+else
+    source $backend_dir/../common/libbackend.sh
+fi
+
+if [ "x${BUILD_PROFILE}" == "xintel" ]; then
+    EXTRA_PIP_INSTALL_FLAGS+=" --upgrade --index-strategy=unsafe-first-match"
+fi
+
+PYTHON_VERSION="3.12"
+PYTHON_PATCH="12"
+PY_STANDALONE_TAG="20251120"
+
+installRequirements
--- a/backend/python/nemo/protogen.sh
+++ b/backend/python/nemo/protogen.sh
@@ -0,0 +1,11 @@
+#!/bin/bash
+set -e
+
+backend_dir=$(dirname $0)
+if [ -d $backend_dir/common ]; then
+    source $backend_dir/common/libbackend.sh
+else
+    source $backend_dir/../common/libbackend.sh
+fi
+
+python3 -m grpc_tools.protoc -I../.. -I./ --python_out=. --grpc_python_out=. backend.proto
--- a/backend/python/nemo/requirements-cpu.txt
+++ b/backend/python/nemo/requirements-cpu.txt
@@ -0,0 +1,3 @@
+--extra-index-url https://download.pytorch.org/whl/cpu
+torch
+nemo_toolkit[asr]
--- a/backend/python/nemo/requirements-cublas12.txt
+++ b/backend/python/nemo/requirements-cublas12.txt
@@ -0,0 +1,3 @@
+--extra-index-url https://download.pytorch.org/whl/cu128
+torch
+nemo_toolkit[asr]
--- a/backend/python/nemo/requirements-cublas13.txt
+++ b/backend/python/nemo/requirements-cublas13.txt
@@ -0,0 +1,3 @@
+--extra-index-url https://download.pytorch.org/whl/cu130
+torch
+nemo_toolkit[asr]
--- a/backend/python/nemo/requirements-hipblas.txt
+++ b/backend/python/nemo/requirements-hipblas.txt
@@ -0,0 +1,3 @@
+--extra-index-url https://download.pytorch.org/whl/rocm6.3
+torch
+nemo_toolkit[asr]
--- a/backend/python/nemo/requirements-intel.txt
+++ b/backend/python/nemo/requirements-intel.txt
@@ -0,0 +1,3 @@
+--extra-index-url https://download.pytorch.org/whl/xpu
+torch
+nemo_toolkit[asr]
--- a/backend/python/nemo/requirements-l4t12.txt
+++ b/backend/python/nemo/requirements-l4t12.txt
@@ -0,0 +1,3 @@
+--extra-index-url https://pypi.jetson-ai-lab.io/jp6/cu129/
+torch
+nemo_toolkit[asr]
--- a/backend/python/nemo/requirements-l4t13.txt
+++ b/backend/python/nemo/requirements-l4t13.txt
@@ -0,0 +1,3 @@
+--extra-index-url https://download.pytorch.org/whl/cu130
+torch
+nemo_toolkit[asr]
--- a/backend/python/nemo/requirements-mps.txt
+++ b/backend/python/nemo/requirements-mps.txt
@@ -0,0 +1,2 @@
+torch
+nemo_toolkit[asr]
--- a/backend/python/nemo/requirements.txt
+++ b/backend/python/nemo/requirements.txt
@@ -0,0 +1,5 @@
+grpcio==1.71.0
+protobuf
+certifi
+packaging==24.1
+setuptools
--- a/backend/python/nemo/run.sh
+++ b/backend/python/nemo/run.sh
@@ -0,0 +1,9 @@
+#!/bin/bash
+backend_dir=$(dirname $0)
+if [ -d $backend_dir/common ]; then
+    source $backend_dir/common/libbackend.sh
+else
+    source $backend_dir/../common/libbackend.sh
+fi
+
+startBackend $@
--- a/backend/python/nemo/test.py
+++ b/backend/python/nemo/test.py
@@ -0,0 +1,99 @@
+"""
+Tests for the NEMO Toolkit ASR gRPC backend.
+"""
+import unittest
+import subprocess
+import time
+import os
+import tempfile
+import shutil
+import backend_pb2
+import backend_pb2_grpc
+
+import grpc
+
+# Skip heavy transcription test in CI (model download + inference)
+SKIP_ASR_TESTS = os.environ.get("SKIP_ASR_TESTS", "false").lower() == "true"
+
+
+class TestBackendServicer(unittest.TestCase):
+    def setUp(self):
+        self.service = subprocess.Popen(["python3", "backend.py", "--addr", "localhost:50051"])
+        time.sleep(15)
+
+    def tearDown(self):
+        self.service.terminate()
+        self.service.wait()
+
+    def test_server_startup(self):
+        try:
+            self.setUp()
+            with grpc.insecure_channel("localhost:50051") as channel:
+                stub = backend_pb2_grpc.BackendStub(channel)
+                response = stub.Health(backend_pb2.HealthMessage())
+                self.assertEqual(response.message, b'OK')
+        except Exception as err:
+            print(err)
+            self.fail("Server failed to start")
+        finally:
+            self.tearDown()
+
+    def test_load_model(self):
+        try:
+            self.setUp()
+            with grpc.insecure_channel("localhost:50051") as channel:
+                stub = backend_pb2_grpc.BackendStub(channel)
+                response = stub.LoadModel(backend_pb2.ModelOptions(Model="nvidia/parakeet-tdt-0.6b-v3"))
+                self.assertTrue(response.success, response.message)
+                self.assertEqual(response.message, "Model loaded successfully")
+        except Exception as err:
+            print(err)
+            self.fail("LoadModel service failed")
+        finally:
+            self.tearDown()
+
+    @unittest.skipIf(SKIP_ASR_TESTS, "ASR transcription test skipped (SKIP_ASR_TESTS=true)")
+    def test_audio_transcription(self):
+        temp_dir = tempfile.mkdtemp()
+        audio_file = os.path.join(temp_dir, 'audio.wav')
+        try:
+            # Download a sample audio file for testing
+            url = "https://audio-samples.github.io/samples/mp3/crowd-cheering-and-applause-sound-effect.mp3"
+            result = subprocess.run(
+                ["wget", "-q", url, "-O", audio_file],
+                capture_output=True,
+                text=True,
+                timeout=30,
+            )
+            if result.returncode != 0:
+                self.skipTest(f"Could not download sample audio: {result.stderr}")
+            if not os.path.exists(audio_file):
+                self.skipTest("Sample audio file not found after download")
+
+            self.setUp()
+            with grpc.insecure_channel("localhost:50051") as channel:
+                stub = backend_pb2_grpc.BackendStub(channel)
+                load_response = stub.LoadModel(backend_pb2.ModelOptions(Model="nvidia/parakeet-tdt-0.6b-v3"))
+                self.assertTrue(load_response.success, load_response.message)
+
+                transcript_response = stub.AudioTranscription(
+                    backend_pb2.TranscriptRequest(dst=audio_file)
+                )
+                self.assertIsNotNone(transcript_response)
+                self.assertIsNotNone(transcript_response.text)
+                self.assertGreaterEqual(len(transcript_response.segments), 0)
+                all_text = ""
+                for segment in transcript_response.segments:
+                    all_text += segment.text
+                print(f"Transcription result: {all_text}")
+                self.assertIn("big", all_text)
+                if transcript_response.segments:
+                    self.assertIsNotNone(transcript_response.segments[0].text)
+        finally:
+            self.tearDown()
+            if os.path.exists(temp_dir):
+                shutil.rmtree(temp_dir)
+
+
+if __name__ == '__main__':
+    unittest.main()
--- a/backend/python/nemo/test.sh
+++ b/backend/python/nemo/test.sh
@@ -0,0 +1,11 @@
+#!/bin/bash
+set -e
+
+backend_dir=$(dirname $0)
+if [ -d $backend_dir/common ]; then
+    source $backend_dir/common/libbackend.sh
+else
+    source $backend_dir/../common/libbackend.sh
+fi
+
+runUnittests