Merge pull request #17 from projecte-aina/tests

Refactor the api based on mvc structure

.github/workflows/tests.yml

@@ -0,0 +1,60 @@
+name: Python Unit Tests
+
+
+on:
+  workflow_dispatch:
+  pull_request:
+    branches:
+    - main
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    steps:
+      -
+        name: Checkout
+        uses: actions/checkout@v3
+
+      - name: Set up Python
+        # This is the version of the action for setting up Python, not the Python version.
+        uses: actions/setup-python@v5
+        with:
+          # Semantic version range syntax or exact version of a Python version
+          python-version: '3.10.12'
+          # Optional - x64 or x86 architecture, defaults to x64
+          architecture: 'x64'
+
+      - name: Display Python version
+        run: python -c "import sys; print(sys.version)"
+
+      - name: Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install pytest httpx pytest-repeat
+          pip install -r requirements.txt
+      
+      - name: install apt dependecies
+        run:  |
+          sudo apt-get update
+          sudo apt-get install -y build-essential autoconf automake libtool pkg-config git wget cmake
+          sudo rm -rf /var/lib/apt/lists/*
+      
+      - name: install espeak-ng
+        run:  |
+          git clone -b dev-ca https://github.com/projecte-aina/espeak-ng
+          pip install --upgrade pip 
+          cd espeak-ng && sudo ./autogen.sh && sudo ./configure --prefix=/usr && sudo make && sudo make install
+
+      - name: install lingua-franca
+        run: pip install git+https://github.com/MycroftAI/lingua-franca.git@5bfd75fe5996fd364102a0eec3f714c9ddc9275c
+
+      - name: install model
+        run:  |
+          wget -q http://share.laklak.eu/model_vits_ca/best_model_8khz.pth -P ./models/vits_ca/
+          mv ./models/vits_ca/best_model_8khz.pth ./models/vits_ca/best_model.pth
+
+      - name: Run tests
+        run: pytest
+
+
+

Dockerfile

@@ -33,4 +33,4 @@ RUN wget -q http://share.laklak.eu/model_vits_ca/best_model_8khz.pth -P /app/mod
 RUN mv /app/models/vits_ca/best_model_8khz.pth /app/models/vits_ca/best_model.pth
 COPY . .
 
-ENTRYPOINT python server/server.py --speech_speed ${SPEECH_SPEED} --mp_workers ${MP_WORKERS} --use_cuda ${USE_CUDA} --use_mp ${USE_MP}
+ENTRYPOINT python main.py --speech_speed ${SPEECH_SPEED} --mp_workers ${MP_WORKERS} --use_cuda ${USE_CUDA} --use_mp ${USE_MP}

Dockerfile.dev

@@ -0,0 +1,35 @@
+FROM python:3.10.12-slim
+# RUN apt-get update && apt-get install -y --no-install-recommends wget gcc g++ make python3 python3-dev python3-pip python3-venv python3-wheel espeak espeak-ng libsndfile1-dev && rm -rf /var/lib/apt/lists/*
+
+# Install required packages for building eSpeak and general utilities
+RUN apt-get update && apt-get install -y \
+        build-essential \
+        autoconf \
+        automake \
+        libtool \
+        pkg-config \
+        git \ 
+        wget \
+        cmake \ 
+    && rm -rf /var/lib/apt/lists/*
+
+RUN git clone -b dev-ca https://github.com/projecte-aina/espeak-ng
+
+RUN pip install --upgrade pip && \
+ cd espeak-ng && \
+ ./autogen.sh && \
+ ./configure --prefix=/usr && \
+ make && \
+ make install
+
+RUN pip install git+https://github.com/MycroftAI/lingua-franca.git@5bfd75fe5996fd364102a0eec3f714c9ddc9275c
+
+WORKDIR /app
+# RUN wget -q http://share.laklak.eu/model_vits_ca/best_model_8khz.pth -P /app/models/vits_ca/
+# RUN mv /app/models/vits_ca/best_model_8khz.pth /app/models/vits_ca/best_model.pth
+
+COPY ./requirements.txt /app
+RUN python -m pip install --upgrade pip
+RUN python -m pip install --no-cache-dir -r requirements.txt
+
+ENTRYPOINT python main.py --speech_speed ${SPEECH_SPEED} --mp_workers ${MP_WORKERS} --use_cuda ${USE_CUDA} --use_mp ${USE_MP} --show_details True --reload

Dockerfile.test

@@ -0,0 +1,34 @@
+FROM python:3.10.12-slim
+# RUN apt-get update && apt-get install -y --no-install-recommends wget gcc g++ make python3 python3-dev python3-pip python3-venv python3-wheel espeak espeak-ng libsndfile1-dev && rm -rf /var/lib/apt/lists/*
+
+# Install required packages for building eSpeak and general utilities
+RUN apt-get update && apt-get install -y \
+        build-essential \
+        autoconf \
+        automake \
+        libtool \
+        pkg-config \
+        git \ 
+        wget \
+        cmake \ 
+    && rm -rf /var/lib/apt/lists/*
+
+RUN git clone -b dev-ca https://github.com/projecte-aina/espeak-ng
+
+RUN pip install --upgrade pip && \
+ cd espeak-ng && \
+ ./autogen.sh && \
+ ./configure --prefix=/usr && \
+ make && \
+ make install
+
+RUN pip install git+https://github.com/MycroftAI/lingua-franca.git@5bfd75fe5996fd364102a0eec3f714c9ddc9275c
+
+WORKDIR /app
+
+COPY ./requirements.txt /app
+RUN python -m pip install --upgrade pip
+RUN python -m pip install --no-cache-dir -r requirements.txt
+RUN pip install pytest httpx pydub pytest-repeat
+
+ENTRYPOINT pytest

Makefile

@@ -1,8 +1,18 @@
 deploy:
 	docker compose --env-file .env up -d --build
+deploy-d:
+	docker compose --env-file .env up --build
 deploy-gpu: 
 	docker compose -f docker-compose-gpu.yml --env-file .env up -d --build
+dev:
+	docker compose -f docker-compose-dev.yml up --build
+tests:
+	docker compose -f docker-compose-test.yml up --build
 undeploy:
 	docker compose down
 stop:
-	docker compose stop
+	docker compose stop
+
+
+act-run-tests:
+	gh act -j test -W '.github/workflows/tests.yml'

README.md

@@ -92,6 +92,22 @@ docker run --shm-size=1gb -p 8080:8000 tts-api
 The default entrypoint puts the web interface to `http://0.0.0.0:8080/`.
 
 
+## Developer Mode
+You can run this api with docker with reload mode that will let you watch you local changes on api.<br>
+To run in dev mode run the following command.
+
+```bash
+make dev
+```
+
+> [!NOTE]
+> The model **best_model.pth** is requiered, you have to download by yourself.
+```bash
+wget -q http://share.laklak.eu/model_vits_ca/best_model_8khz.pth -P models/vits_ca/
+```
+```bash
+mv models/vits_ca/best_model_8khz.pth models/vits_ca/best_model.pth
+```
 
 ## REST API Endpoints
 

docker-compose-dev.yml

@@ -0,0 +1,18 @@
+version: '3.9'
+services:
+  server:
+    build:
+      context: .
+      dockerfile: Dockerfile.dev
+    container_name: fastapi-dev
+    environment:
+      - SPEECH_SPEED=${SPEECH_SPEED-1.0}
+      - MP_WORKERS=${MP_WORKERS-4}
+      - USE_MP=False
+    restart: always
+    volumes:
+      - .:/app
+
+    ports:
+      - '8080:8000'
+    shm_size: ${SHM_SIZE-2gb}

docker-compose-test.yml

@@ -0,0 +1,14 @@
+version: '3.9'
+services:
+  server:
+    build:
+      context: .
+      dockerfile: Dockerfile.test
+    container_name: fastapi-test
+    restart: always
+    volumes:
+      - .:/app
+
+    ports:
+      - '8080:8000'
+    shm_size: ${SHM_SIZE-2gb}

main.py

@@ -0,0 +1,153 @@
+from pathlib import Path
+from TTS.utils.manage import ModelManager
+from lingua_franca import load_language # Lingua franca
+
+import argparse
+import uvicorn
+import torch
+import multiprocessing as mp
+import sys
+import os
+
+from server import create_app
+from server.utils.argparse import MpWorkersAction
+from server.utils.utils import update_config
+from server.utils.utils import update_config
+
+# Set global paths
+# Determine the current script's directory and set up paths related to the model
+path = Path(__file__).parent / "server" /".models.json"
+path_dir = os.path.dirname(path)
+
+# Initialize the model manager with the aforementioned path
+manager = ModelManager(path)
+
+# Set the relative paths for the default TTS model and its associated configuration
+models_path_rel = '../models/vits_ca'
+model_ca = os.path.join(path_dir, models_path_rel, 'best_model.pth')
+config_ca = os.path.join(path_dir, models_path_rel, 'config.json')
+
+# Load lingua franca language
+load_language('ca-es')
+
+def create_argparser():
+    def convert_boolean(x):
+        return x.lower() in ["true", "1", "yes"]
+
+    # Create an argument parser to handle command-line arguments
+    # The parser setup seems incomplete and might be continued in the next section of the code.
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "--list_models",
+        type=convert_boolean,
+        nargs="?",
+        const=True,
+        default=False,
+        help="list available pre-trained tts and vocoder models."
+    )
+    parser.add_argument(
+        "--model_name",
+        type=str,
+        default="tts_models/en/ljspeech/tacotron2-DDC",
+        help="Name of one of the pre-trained tts models in format <language>/<dataset>/<model_name>",
+    )
+    parser.add_argument("--vocoder_name", type=str, default=None, help="name of one of the released vocoder models.")
+    # Args for running custom models
+    parser.add_argument(
+        "--config_path",
+        default=config_ca,
+        type=str,
+        help="Path to model config file."
+    )
+    parser.add_argument(
+        "--model_path",
+        type=str,
+        default=model_ca,
+        help="Path to model file.",
+    )
+    parser.add_argument(
+        "--vocoder_path",
+        type=str,
+        help="Path to vocoder model file. If it is not defined, model uses GL as vocoder. Please make sure that you installed vocoder library before (WaveRNN).",
+        default=None,
+    )
+    parser.add_argument("--vocoder_config_path", type=str, help="Path to vocoder model config file.", default=None)
+    parser.add_argument("--speakers_file_path", type=str, help="JSON file for multi-speaker model.", default=None)
+    parser.add_argument("--port", type=int, default=8000, help="port to listen on.")
+    parser.add_argument("--host", type=str, default="0.0.0.0", help="host ip to listen.")
+    parser.add_argument("--use_mp", type=convert_boolean, default=False, nargs='?', const=True, help="true to use Python multiprocessing.")
+    parser.add_argument("--use_cuda", type=convert_boolean, default=False, nargs='?', const=False, help="true to use CUDA.")
+    parser.add_argument("--mp_workers", action=MpWorkersAction ,type=int, default=mp.cpu_count(), nargs='?', const=mp.cpu_count(), help="number of CPUs used for multiprocessing")
+    parser.add_argument("--debug", type=convert_boolean, default=False, help="true to enable Flask debug mode.")
+    parser.add_argument("--show_details", type=convert_boolean, default=False, help="Generate model detail page.")
+    parser.add_argument("--speech_speed", type=float, default=1.0, nargs='?', const=1.0, help="Change speech speed.")
+    parser.add_argument("--reload", type=bool, action=argparse.BooleanOptionalAction, default=False, help="Reload on changes")
+    return parser
+
+
+# parse the args
+args = create_argparser().parse_args()
+print("args =========", args)
+# update in-use models to the specified released models.
+model_path = None
+config_path = None
+speakers_file_path = None
+vocoder_path = None
+vocoder_config_path = None
+# new_speaker_ids = None
+# use_aliases = None
+
+# CASE1: list pre-trained TTS models
+if args.list_models:
+    manager.list_models()
+    sys.exit()
+
+# CASE2: load pre-trained model paths
+if args.model_name is not None and not args.model_path:
+    model_path, config_path, model_item = manager.download_model(args.model_name)
+    args.vocoder_name = model_item["default_vocoder"] if args.vocoder_name is None else args.vocoder_name
+
+if args.vocoder_name is not None and not args.vocoder_path:
+    vocoder_path, vocoder_config_path, _ = manager.download_model(args.vocoder_name)
+
+# CASE3: set custom model paths
+if args.model_path is not None:
+    model_path = args.model_path
+    config_path = args.config_path
+    speakers_file_path = args.speakers_file_path
+    speaker_ids_path = os.path.join(path_dir, models_path_rel, 'speaker_ids.json')
+
+if args.vocoder_path is not None:
+    vocoder_path = args.vocoder_path
+    vocoder_config_path = args.vocoder_config_path
+
+# CASE4: change speaker speed
+if args.speech_speed != 1.0:
+    update_config(config_path, args.speech_speed)
+
+
+app = create_app(
+        model_path = model_path, 
+        config_path = config_path, 
+        speakers_file_path = speakers_file_path, 
+        vocoder_path = vocoder_path, 
+        vocoder_config_path = vocoder_config_path, 
+        speaker_ids_path = speaker_ids_path, 
+        speech_speed = args.speech_speed, 
+        mp_workers = args.mp_workers, 
+        use_cuda = args.use_cuda, 
+        use_mp = args.use_mp,
+        show_details=args.show_details,
+        args=args
+    )
+
+
+
+def main():
+    uvicorn.run('main:app', host=args.host, port=args.port, reload=args.reload)
+
+if __name__ == "__main__":
+    torch.set_num_threads(1)
+    torch.set_grad_enabled(False)
+    mp.set_start_method("fork")
+    main()