restructure_for_cloud (#3)

- implement a presentable working version of translator_Server

Reviewed-on: https://gitea.pstruebi.xyz/auracaster/multilang-translator-local/pulls/3

.vscode/launch.json

@@ -10,7 +10,8 @@
             "type": "debugpy",
             "request": "launch",
             "program": "${file}",
-            "console": "integratedTerminal"
+            "console": "integratedTerminal",
+            "justMyCode": true
         }
     ]
 }

.vscode/tasks.json

@@ -13,6 +13,12 @@
             "label": "pip install -e bumble",
             "type": "shell",
             "command": "./venv/bin/python -m pip install -e ../bumble --config-settings editable_mode=compat"
+        },
+        {
+            "label": "pip install -e auracast",
+            "type": "shell",
+            "command": "./venv/bin/python -m pip install -e ../bumble-auracast --config-settings editable_mode=compat",
+            "problemMatcher": []
         }
     ]
 }

multilang_translator/config.py

@@ -1,39 +0,0 @@
-import os
-
-ANNOUNCEMENT_DIR = os.path.join(os.path.dirname(__file__), 'announcements')
-VENV_DIR = os.path.join(os.path.dirname(__file__), '../venv')
-PIPER_EXE_PATH = f'{VENV_DIR}/bin/piper'
-FRAME_DUR_MS = 10
-SAMPLING_RATE_HZ = int(16e3)
-BITRATE_BPS = int(32e3)
-
-def mk_filename_lc3(lang=''):
-    return f"{ANNOUNCEMENT_DIR}/announcement_{lang}_{SAMPLING_RATE_HZ//1000}_{FRAME_DUR_MS}_{BITRATE_BPS//1000}.lc3"
-
-LANG_CONFIG = {
-    "de": {
-        "filepath_wav": f"{ANNOUNCEMENT_DIR}/announcement_de.wav",
-        "filepath_wav_resamp": f"{ANNOUNCEMENT_DIR}/announcement_de_resamp.wav",
-        "tts": 'de_DE-kerstin-low',
-    },
-    "en": {
-        "filepath_wav": f"{ANNOUNCEMENT_DIR}/announcement_en.wav",
-        "filepath_wav_resamp": f"{ANNOUNCEMENT_DIR}/announcement_en_resamp.wav",
-        "tts": 'en_US-lessac-medium'
-    },
-    "fr": {
-        "filepath_wav": f"{ANNOUNCEMENT_DIR}/announcement_fr.wav",
-        "filepath_wav_resamp": f"{ANNOUNCEMENT_DIR}/announcement_fr_resamp.wav",
-        "tts": 'fr_FR-siwis-medium'
-    },
-    # "es": {
-    #     "filepath_wav": f"{ANNOUNCEMENT_DIR}/announcement_es.wav",
-    #     "tts": 'es_ES-sharvard-medium'
-    # },
-    # "it": {
-    #     "filepath_wav": f"{ANNOUNCEMENT_DIR}/announcement_it.wav",
-    #     "tts": 'it_IT-paola-medium'
-    # }
-}
-
-os.makedirs(ANNOUNCEMENT_DIR, exist_ok=True)

multilang_translator/main.py

@@ -1,131 +0,0 @@
-# -*- coding: utf-8 -*-
-"""
-list prompt example
-"""
-from __future__ import print_function, unicode_literals
-
-from dataclasses import asdict
-import asyncio
-from copy import copy
-import time
-import logging as log
-import aioconsole
-
-from utils import resample
-from translator import llm_translator, test_content
-from text_to_speech import text_to_speech
-from encode import encode_lc3
-from auracast import multicast_control
-from auracast import auracast_config
-from config import LANG_CONFIG, SAMPLING_RATE_HZ
-from translator.test_content import TESTSENTENCE
-
-# TODO: look for a end to end translation solution
-
-def transcribe():
-    pass # TODO: Implement transcribing input audio e.g. with whisper
-
-
-def syntesize_resample(text, tts_model, file_wav, file_wav_resamp):
-    audio_dur = text_to_speech.synthesize(text, tts_model, file_wav)
-    resample.resample_file(file_wav, file_wav_resamp, target_rate=SAMPLING_RATE_HZ)
-    return audio_dur
-
-
-def translate_from_german(text_de, model):
-    config = copy(LANG_CONFIG)
-    base_lang = "de"
-
-    file = config[base_lang]["filepath_wav"]
-    file_resamp = config[base_lang]['filepath_wav_resamp']
-    tts_json = {}
-
-    for key, val in config.items():
-        if key == base_lang:
-            text = text_de
-        else:
-            text = llm_translator.translate_de_to_x(text_de, key, model=model)
-        
-        log.info('%s', text)
-        file = val['filepath_wav']
-        file_resamp = val['filepath_wav_resamp']
-        tts_json[key] = syntesize_resample(text, val['tts'], file, file_resamp)
-
-    return tts_json
-
-
-async def announcement_from_german_text(caster:multicast_control.Multicaster, text_de):
-
-    tts_json = translate_from_german(text_de, model = 'llama3.2:3b-instruct-q4_0')
-
-    start = time.time()
-    await caster.init_audio()
-    caster.start_streaming()
-
-    log.info("Starting all broadcasts took %s s", round(time.time() - start, 3))
-
-
-async def command_line_ui(caster: multicast_control.Multicaster):
-    while True:
-        # make a list of all available testsentence
-        sentence_list = list(asdict(TESTSENTENCE).values())
-
-        prompt = "Enter your Announcement|quit or choose:] > \n"
-        prompt += "\n".join([f"{i}: {s}" for i,s in enumerate(sentence_list)])
-        prompt += "\n"
-        command = await aioconsole.ainput(prompt)
-    
-        if command.strip().lower() == "quit":
-            print("👋 Exiting...")
-            if caster.device:
-                caster.stop_streaming()
-                await caster.shutdown()
-            break  # Exit loop
-        elif command.strip() == '':
-            print('Nothing to Announce')
-        # Check if command is a single number
-        elif command.strip().isdigit():
-            ind = int(command.strip())
-            await announcement_from_german_text(caster, sentence_list[ind])
-            await asyncio.wait([caster.streamer.task])
-        # Interpret the command as announcement
-        else:
-            await announcement_from_german_text(caster, command)
-            await asyncio.wait([caster.streamer.task])
-
-async def main():
-    log.basicConfig(
-        level=log.INFO,
-        format='%(module)s.py:%(lineno)d %(levelname)s: %(message)s'
-    )
-
-    global_conf = auracast_config.global_base_config
-    #global_conf.transport='serial:/dev/serial/by-id/usb-SEGGER_J-Link_001057705357-if02,1000000,rtscts' # transport for nrf54l15dk
-    global_conf.transport='serial:/dev/serial/by-id/usb-ZEPHYR_Zephyr_HCI_UART_sample_81BD14B8D71B5662-if00,115200,rtscts' #nrf52dongle hci_uart usb cdc
-
-    big_conf = [ 
-        auracast_config.broadcast_de,
-        auracast_config.broadcast_en,
-        auracast_config.broadcast_fr,
-        #auracast_config.broadcast_es,
-        #auracast_config.broadcast_it,
-        ]
-    files = [v['filepath_wav_resamp'] for v in LANG_CONFIG.values()]
-    for i, conf in enumerate(big_conf):
-        conf.loop_wav = False
-        conf.audio_source = f'file:{files[i]}'
-
-    caster = multicast_control.Multicaster(global_conf, big_conf)
-    await caster.init_broadcast()
-    
-    #await announcement_from_german_text(caster, test_content.TESTSENTENCE_DE_HELLO)
-
-    await command_line_ui(caster)
-    #await asyncio.wait([caster.streamer.task])
-
-if __name__ == '__main__':
-    asyncio.run(main())
-
-    # TODO: integrate this in the LANG_CONFIG dict, better: make a hierachy of dataclasses
-    # TODO: remove the nececcity for files
-    # TODO: add support for multiple radios

multilang_translator/text_to_speech/text_to_speech.py

@@ -1,42 +0,0 @@
-import os
-import subprocess
-import time
-import json
-import logging as log
-from multilang_translator import config
-
-TTS_DIR = os.path.join(os.path.dirname(__file__))
-
-def synthesize(text, model="en_US-lessac-medium", output_file="out.wav"):
-
-    pwd = os.getcwd()
-    os.chdir(TTS_DIR)
-    start = time.time()
-    
-    ret = subprocess.run( # TODO: wrap this whole thing in a class and open a permanent instance of the model 
-        [config.PIPER_EXE_PATH, '--model', model, '--output_file', output_file], 
-        input=text.encode('utf-8'),
-        capture_output=True
-        )
-    log.info('%s', ret.stdout)
-    log.info('%s', ret.stderr)
-
-    assert ret.returncode == 0, 'Piper returncode was not 0.'
-
-    log.info("Running piper for model %s took %s s", model, round(time.time() - start, 3))
-
-    with open (f'{model}.onnx.json') as f: # TODO: wrap everything into a class, store the json permanentl
-        model_json = json.load(f)
-
-    os.chdir(pwd)
-    return model_json
-
-if __name__ == '__main__':
-    import logging
-
-    logging.basicConfig(
-        level=logging.INFO,
-        format='%(module)s.py:%(lineno)d %(levelname)s: %(message)s'
-    )
-
-    synthesize('Hello World')

multilang_translator/translator/__init__.py

@@ -1,2 +0,0 @@
-from .credentials import *
-from .syspromts import *

multilang_translator/translator/credentials.py

@@ -1,2 +0,0 @@
-BASE_URL='https://ollama.hinterwaldner.duckdns.org'
-TOKEN = 'sk-17124cb84df14cc6ab2d9e17d0724d13'

multilang_translator/translator/llm_translator.py

@@ -1,61 +0,0 @@
-import time
-import requests
-import json
-import logging as log
-import time
-import ollama
-
-from multilang_translator.translator import credentials
-from multilang_translator.translator import syspromts
-from multilang_translator.translator import test_content
-
-# ollama.create( # TODO: create models on startup 
-#     model='example', 
-#     from_='llama3.2', system="You are Mario from Super Mario Bros."
-#     )
-
-def query_model(model, query):
-    url = f'{credentials.BASE_URL}/api/chat/completions'
-    headers = {
-        'Authorization': f'Bearer {credentials.TOKEN}',
-    }
-    payload = {
-        'model': model,
-        'messages': [{'role': 'user', 'content': query}],
-    }
-    start = time.time()
-    response = requests.post(url, headers=headers, json=payload)
-    log.info("Translating the text took %s s", round(time.time() - start, 2))
-    return response.json()
-
-
-def translate_de_to_x(text:str, target_language: str, model='llama3.2:3b-instruct-q4_0'): # remember to use instruct models
-    start=time.time()
-    s = getattr(syspromts, f"TRANSLATOR_DE_{target_language.upper()}")
-    response = ollama.chat(
-        model = model,
-        messages = [
-            {'role': 'system', 'content': s},
-            {'role': 'user', 'content': text}
-            ],
-        )
-    log.info('Running the translator to %s took %s s', target_language, round(time.time() - start, 3))
-    return response['message']['content']
-
-if __name__ == "__main__":
-    import time
-
-    start=time.time()
-    response = translate_de_to_x('Der Zug ist da.', target_language='en', model='llama3.2:1b-instruct-q4_0')
-    print("Query took", time.time() - start) 
-    print(response)
-
-    start=time.time()
-    response = translate_de_to_x(test_content.TESTSENTENCE_DE_RAINBOW, target_language='en')
-    print("query took", time.time() - start) 
-    print(response)
-
-    start=time.time()
-    response = translate_de_to_x(test_content.TESTSENTENCE_DE_RAINBOW, target_language='fr')
-    print("query took", time.time() - start) 
-    print(response)

multilang_translator/translator/syspromts.py

@@ -1,4 +0,0 @@
-TRANSLATOR_DE_EN = 'Du bist ein Übersetzer. Übersetze die folgende Satz aus dem Deutschen ins Englische. Antworte nur mit der übersetzten Satz.\n'
-TRANSLATOR_DE_FR = 'Du bist ein Übersetzer. Übersetze die folgende Satz aus dem Deutschen ins Französische. Antworte nur mit der übersetzten Satz.\n'
-TRANSLATOR_DE_ES = 'Du bist ein Übersetzer. Übersetze die folgende Satz aus dem Deutschen ins Spanische. Antworte nur mit der übersetzten Satz.\n'
-TRANSLATOR_DE_IT = 'Du bist ein Übersetzer. Übersetze die folgende Satz aus dem Deutschen ins Italienische. Antworte nur mit der übersetzten Satz.\n'

multilang_translator/utils/resample.py

@@ -1,27 +0,0 @@
-# resample .wav source to target sampling rate
-import logging as log
-import time
-import os
-import librosa
-import soundfile as sf
-
-
-def resample_file(filename, out_filename, target_rate=int(24e3)):
-    start=time.time()
-    # Load the original audio file
-    audio, rate = librosa.load(filename)
-
-    if rate == target_rate: # Nothing to do 
-        sf.write(out_filename, audio, target_rate)
-        return
-
-    # Convert the sample rate to 24 kHz
-    resampled_audio = librosa.resample(audio, orig_sr=rate, target_sr=target_rate)
-
-    # Save the resampled audio as a new .wav file
-    sf.write(out_filename, resampled_audio, target_rate)
-    
-    log.info("Resampling of %s took  %s s", os.path.basename(filename), round(time.time() - start, 3))
-
-if __name__ == "__main__":
-    resample_file('text_to_speech/welcome.wav', 'text_to_speech/welcome_resampled.wav')

pyproject.toml

@@ -5,10 +5,12 @@ version = '0.1'
 
 dependencies = [
     "auracast @git+https://git@gitea.pstruebi.xyz/auracaster/bumble-auracast",
-    "requests",
-    "ollama",
-    "aioconsole",
-    "piper-tts==1.2.0"
+    "requests==2.32.3",
+    "ollama==0.4.7",
+    "aioconsole==0.8.1",
+    "fastapi==0.115.11",
+    "uvicorn==0.34.0",
+    "aiohttp==3.9.3",
 ]
 
 [project.optional-dependencies]
@@ -16,6 +18,11 @@ test = [
     "pytest >= 8.2",
 ]
 
+[tool.poetry.group.tts.dependencies]
+piper-phonemize = "==1.1.0"
+piper-tts = "==1.2.0"
+
+
 [tool.pytest.ini_options]
 addopts = [
     "--import-mode=importlib","--count=1","-s","-v" 

run_local.sh

@@ -1,3 +0,0 @@
-#!/bin/bash
-source ./venv/bin/activate
-python ./multilang_translator/main.py

src/multilang_translator/main_cloud.py

@@ -0,0 +1,89 @@
+from typing import List
+import time
+import asyncio
+
+import logging as log
+
+from auracast import multicast_client
+from auracast import auracast_config
+
+import voice_client
+import voice_models
+
+from multilang_translator import translator_config
+from multilang_translator.translator import llm_translator
+import voice_client.tts_client
+import voice_models.request_models
+
+
+async def announcement_from_german_text(
+        config: translator_config.TranslatorConfigGroup,
+        text_de
+        ):
+    base_lang = "deu"
+
+    audio_data_dict = {}
+    for i, big in enumerate(config.bigs):
+        if big.language == base_lang:
+            text = text_de
+        else:
+            text = llm_translator.translate_de_to_x(
+                text_de,
+                big.language,
+                model=big.translator_llm,
+                client = big.llm_client,
+                host=big.llm_host_url,
+                token=big.llm_host_token
+            )
+
+        log.info('%s', text)
+        request_data = voice_models.request_models.SynthesizeRequest(
+            text=text,
+            target_sample_rate=config.auracast_sampling_rate_hz,
+            framework=big.tts_system, 
+            model=big.tts_model,
+            return_lc3=True
+        )
+        start = time.time()
+        lc3_audio = voice_client.tts_client.request_synthesis(
+            request_data
+        )
+        log.info('Voice synth took %s', time.time() - start)
+        audio_data_dict[big.language] = lc3_audio.decode('latin-1') # TODO: should be .hex in the future
+
+    await multicast_client.send_audio(
+        audio_data_dict
+    )
+
+
+async def main():
+    log.basicConfig(
+        level=log.INFO,
+        format='%(module)s.py:%(lineno)d %(levelname)s: %(message)s'
+    )
+
+    config = translator_config.TranslatorConfigGroup(
+        bigs=[
+            translator_config.TranslatorConfigDe(),
+            translator_config.TranslatorConfigEn(),
+            translator_config.TranslatorConfigFr(),
+        ]
+    )
+
+    config.transport='serial:/dev/serial/by-id/usb-ZEPHYR_Zephyr_HCI_UART_sample_81BD14B8D71B5662-if00,115200,rtscts' #nrf52dongle hci_uart usb cdc
+
+    for conf in config.bigs:
+        conf.loop = False
+        conf.llm_client = 'openwebui' # comment out for local llm
+        conf.llm_host_url = 'https://ollama.pstruebi.xyz' 
+        conf.llm_host_token = 'sk-17124cb84df14cc6ab2d9e17d0724d13'
+
+    await multicast_client.init(
+        config
+    )
+
+    await announcement_from_german_text(config, 'Hello')
+
+
+if __name__ == '__main__':
+    asyncio.run(main())

src/multilang_translator/main_local.py

@@ -0,0 +1,136 @@
+from typing import List
+from dataclasses import asdict
+import asyncio
+import time
+import logging as log
+import aioconsole
+
+from auracast import multicast_control
+from auracast import auracast_config
+from voice_provider import text_to_speech
+
+from multilang_translator import translator_config
+from multilang_translator.translator import llm_translator
+from multilang_translator.translator.test_content import TESTSENTENCE
+
+# TODO: look for a end to end translation solution
+
+def transcribe():
+    pass # TODO: Implement transcribing input audio e.g. with whisper
+
+
+async def announcement_from_german_text(
+        config: translator_config.TranslatorConfigGroup,
+        caster: multicast_control.Multicaster,
+        text_de
+        ):
+    base_lang = "deu"
+
+    for i, big in enumerate(config.bigs):
+        if big.language == base_lang:
+            text = text_de
+        else:
+            text = llm_translator.translate_de_to_x(
+                text_de,
+                big.language,
+                model=big.translator_llm,
+                client = big.llm_client,
+                host=big.llm_host_url,
+                token=big.llm_host_token
+            )
+
+        log.info('%s', text)
+
+        lc3_audio = text_to_speech.synthesize(
+            text,
+            config.auracast_sampling_rate_hz,
+            big.tts_system,
+            big.tts_model,
+            return_lc3=True
+            )
+        caster.big_conf[i].audio_source = lc3_audio
+
+    start = time.time()
+    caster.start_streaming()
+
+    log.info("Starting all broadcasts took %s s", round(time.time() - start, 3))
+
+
+async def command_line_ui(config: translator_config.TranslatorConfigGroup, translator_conf, caster: multicast_control.Multicaster):
+    while True:
+        # make a list of all available testsentence
+        sentence_list = list(asdict(TESTSENTENCE).values())
+
+        prompt = "Enter your Announcement|quit or choose:] > \n"
+        prompt += "\n".join([f"{i}: {s}" for i,s in enumerate(sentence_list)])
+        prompt += "\n"
+        command = await aioconsole.ainput(prompt)
+
+        if command.strip().lower() == "quit":
+            print("👋 Exiting...")
+            if caster.device:
+                caster.stop_streaming()
+                await caster.shutdown()
+            break  # Exit loop
+        elif command.strip() == '':
+            print('Nothing to Announce')
+        # Check if command is a single number
+        elif command.strip().isdigit():
+            ind = int(command.strip())
+            await announcement_from_german_text(
+                config,
+                caster, 
+                sentence_list[ind])
+            await asyncio.wait([caster.streamer.task])
+        # Interpret the command as announcement
+        else:
+            await announcement_from_german_text(caster, command)
+            await asyncio.wait([caster.streamer.task])
+
+
+async def main():
+    log.basicConfig(
+        level=log.INFO,
+        format='%(module)s.py:%(lineno)d %(levelname)s: %(message)s'
+    )
+
+    config = translator_config.TranslatorConfigGroup(
+        bigs=[
+            translator_config.TranslatorConfigDe(),
+            translator_config.TranslatorConfigEn(),
+            translator_config.TranslatorConfigFr(),
+        ]
+    )
+
+    #config = auracast_config.AuracastGlobalConfig()
+    #config.transport='serial:/dev/serial/by-id/usb-SEGGER_J-Link_001057705357-if02,1000000,rtscts' # transport for nrf54l15dk
+    config.transport='serial:/dev/serial/by-id/usb-ZEPHYR_Zephyr_HCI_UART_sample_81BD14B8D71B5662-if00,115200,rtscts' #nrf52dongle hci_uart usb cdc
+
+    for conf in config.bigs:
+        conf.loop = False
+        conf.llm_client = 'openwebui' # comment out for local llm
+        conf.llm_host_url = 'https://ollama.pstruebi.xyz' 
+        conf.llm_host_token = 'sk-17124cb84df14cc6ab2d9e17d0724d13'
+
+    caster = multicast_control.Multicaster(
+        config, 
+        [big for big in config.bigs]
+        )
+    await caster.init_broadcast()
+
+    # await announcement_from_german_text(
+    #     config,
+    #     caster, 
+    #     test_content.TESTSENTENCE.DE_HELLO
+    #     )
+    # await asyncio.wait([caster.streamer.task])
+    await command_line_ui(
+        config, 
+        [big for big in config.bigs], 
+        caster
+        )
+
+if __name__ == '__main__':
+    asyncio.run(main())
+
+    # TODO: add support for multiple radios

src/multilang_translator/translator/llm_translator.py

@@ -0,0 +1,144 @@
+import time
+import requests
+import json
+import logging as log
+import time
+import ollama
+import aiohttp
+
+from multilang_translator.translator import syspromts
+
+# ollama.create( # TODO: create models on startup 
+#     model='example', 
+#     from_='llama3.2', system="You are Mario from Super Mario Bros."
+#     )
+
+
+def query_openwebui(model, system, query, url, token):
+    url = f'{url}/api/chat/completions'
+    headers = {
+        'Authorization': f'Bearer {token}',
+    }
+    payload = {
+        'model': model,
+        'messages': [
+            {'role': 'system', 'content': system},
+            {'role': 'user', 'content': query}
+        ],
+    }
+    start = time.time()
+    response = requests.post(url, headers=headers, json=payload)
+    log.info("Translating the text took %s s", round(time.time() - start, 2))
+    return response.json()['choices'][0]['message']['content']
+
+
+def query_ollama(model, system, query, host='http://localhost:11434'):
+    client = ollama.Client(
+        host=host,
+    )
+
+    response = client.chat(
+    model = model,
+    messages = [
+        {'role': 'system', 'content': system},
+        {'role': 'user', 'content': query}
+        ],
+    )
+    return response.message.content
+
+
+async def query_openwebui_async(model, system, query, url, token):
+    url = f'{url}/api/chat/completions'
+    headers = {
+        'Authorization': f'Bearer {token}',
+    }
+    payload = {
+        'model': model,
+        'messages': [
+            {'role': 'system', 'content': system},
+            {'role': 'user', 'content': query}
+        ],
+    }
+    start = time.time()
+    async with aiohttp.ClientSession() as session:
+        async with session.post(url, headers=headers, json=payload) as response:
+            response_json = await response.json()
+            log.info("Translating the text took %s s", round(time.time() - start, 2))
+            return response_json['choices'][0]['message']['content']
+
+
+async def query_ollama_async(model, system, query, host='http://localhost:11434'):
+    client = ollama.AsyncClient(
+        host=host,
+    )
+
+    response = await client.chat(
+        model=model,
+        messages=[
+            {'role': 'system', 'content': system},
+            {'role': 'user', 'content': query}
+        ],
+    )
+    return response.message.content
+
+
+def translate_de_to_x( # TODO: use async ollama client later - implenent a translate async function
+        text:str, 
+        target_language: str, 
+        client='ollama', 
+        model='llama3.2:3b-instruct-q4_0', # remember to use instruct models
+        host = None, 
+        token = None
+    ): 
+    start=time.time()
+    s = getattr(syspromts, f"TRANSLATOR_DEU_{target_language.upper()}")
+ 
+    if client == 'ollama':
+        response = query_ollama(model, s, text, host=host)
+    elif client == 'openwebui':
+        response = query_openwebui(model, s, text, url=host, token=token)
+    else: raise NotImplementedError('llm client not implemented')
+
+    log.info('Running the translator to %s took %s s', target_language, round(time.time() - start, 3))
+    return response
+
+async def translate_de_to_x_async(
+        text:str, 
+        target_language: str, 
+        client='ollama', 
+        model='llama3.2:3b-instruct-q4_0', # remember to use instruct models
+        host = None, 
+        token = None
+    ): 
+    start=time.time()
+    s = getattr(syspromts, f"TRANSLATOR_DEU_{target_language.upper()}")
+ 
+    if client == 'ollama':
+        response = await query_ollama_async(model, s, text, host=host)
+    elif client == 'openwebui':
+        response = await query_openwebui_async(model, s, text, url=host, token=token)
+    else: raise NotImplementedError('llm client not implemented')
+
+    log.info('Running the translator to %s took %s s', target_language, round(time.time() - start, 3))
+    return response
+
+
+if __name__ == "__main__":
+    import time
+    from multilang_translator.translator import test_content
+
+
+    start=time.time()
+    response = translate_de_to_x('Der Zug ist da.', target_language='en', model='llama3.2:1b-instruct-q4_0')
+    print("Query took", time.time() - start) 
+    print(response)
+
+    start=time.time()
+    response = translate_de_to_x(test_content.TESTSENTENCE_DE_RAINBOW, target_language='en')
+    print("query took", time.time() - start) 
+    print(response)
+
+    start=time.time()
+    response = translate_de_to_x(test_content.TESTSENTENCE_DE_RAINBOW, target_language='fr')
+    print("query took", time.time() - start) 
+    print(response)

src/multilang_translator/translator/syspromts.py

@@ -0,0 +1,6 @@
+
+# TODO: make this more elegant. this can probably be generated and the base lang be assumed by the llm?
+TRANSLATOR_DEU_ENG = 'Du bist ein Übersetzer. Übersetze die folgende Satz aus dem Deutschen ins Englische. Antworte nur mit der übersetzten Satz.\n'
+TRANSLATOR_DEU_FRA = 'Du bist ein Übersetzer. Übersetze die folgende Satz aus dem Deutschen ins Französische. Antworte nur mit der übersetzten Satz.\n'
+TRANSLATOR_DEU_SPA = 'Du bist ein Übersetzer. Übersetze die folgende Satz aus dem Deutschen ins Spanische. Antworte nur mit der übersetzten Satz.\n'
+TRANSLATOR_DEU_ITA = 'Du bist ein Übersetzer. Übersetze die folgende Satz aus dem Deutschen ins Italienische. Antworte nur mit der übersetzten Satz.\n'

src/multilang_translator/translator/test_content.py

@@ -5,7 +5,6 @@ class TestContent:
     DE_HELLO: str = 'Hallo Welt.'
     DE_GATE_OPENED: str = "Gate 23 ist jetzt geöffnet."
     DE_TRAIN_ARRIVING: str = "Der Zug Nach Wien fährt heute von Gleis 3."
-    DE_SECURITY_CHECKPOINT_OPENING: str = "Sicherheitskontrolle 5 ist jetzt geöffnet. Bitte setzen Sie sich in Bewegung, um Ihre Wartezeit während Sicherungsprüfungen zu minimieren."
+    DE_SECURITY_CHECKPOINT_OPENING: str = "Sicherheitskontrolle 5 ist jetzt geöffnet. Bitte setzen Sie sich in Bewegung, um Ihre Wartezeit während Sicherheitsüberprüfungen zu minimieren."
     DE_RAINBOW: str = 'Der Regenbogen ist ein atmosphärisch-optisches Phänomen, das als kreisbogenförmiges farbiges Lichtband in einer von der Sonne beschienenen Wolke oder Regenwand wahrgenommen wird und ein großes Farbspektrum anzeigt.'
-    DE_WAVE_PARTICLE: str = 'Der Wellen-Teilchen-Dualismus ist eine Konzeption, die postuliert, dass Teilchen sowohl als Wellen auf der Mikroebene verhalten sich und genau bestimme Eigenschaften wie Impuls und Energietrang besaßen.'
 TESTSENTENCE = TestContent()

src/multilang_translator/translator_client/translator_client.py

@@ -0,0 +1,94 @@
+"""
+API client functions for interacting with the Translator API.
+"""
+import requests
+from typing import List, Optional, Dict, Any, Tuple
+from enum import Enum
+
+
+from multilang_translator.translator_models.translator_models import AnnouncementStates, Endpoint, EndpointGroup
+
+
+# This can be overridden through environment variables
+API_BASE_URL = "http://localhost:7999"
+
+def get_groups() -> List[EndpointGroup]:
+    """Get all endpoint groups."""
+    response = requests.get(f"{API_BASE_URL}/groups")
+    response.raise_for_status()
+    return [EndpointGroup.model_validate(group) for group in response.json()]
+
+def get_group(group_id: int) -> Optional[EndpointGroup]:
+    """Get a specific endpoint group by ID."""
+    response = requests.get(f"{API_BASE_URL}/groups/{group_id}")
+    if response.status_code == 404:
+        return None
+    response.raise_for_status()
+    return EndpointGroup.model_validate(response.json())
+
+def create_group(group: EndpointGroup) -> EndpointGroup:
+    """Create a new endpoint group."""
+    # Convert the model to a dict with enum values as their primitive values
+    payload = group.model_dump(mode='json')
+    response = requests.post(f"{API_BASE_URL}/groups", json=payload)
+    response.raise_for_status()
+    return EndpointGroup.model_validate(response.json())
+
+def update_group(group_id: int, updated_group: EndpointGroup) -> EndpointGroup:
+    """Update an existing endpoint group."""
+    # Convert the model to a dict with enum values as their primitive values
+    payload = updated_group.model_dump(mode='json')
+    response = requests.put(f"{API_BASE_URL}/groups/{group_id}", json=payload)
+    response.raise_for_status()
+    return EndpointGroup.model_validate(response.json())
+
+def delete_group(group_id: int) -> None:
+    """Delete an endpoint group."""
+    response = requests.delete(f"{API_BASE_URL}/groups/{group_id}")
+    response.raise_for_status()
+
+def start_announcement(text: str, group_id: int) -> Dict[str, Any]:
+    """
+    Start a new announcement.
+    
+    Args:
+        text: The text content of the announcement
+        group_id: The ID of the endpoint group to send the announcement to
+        
+    Returns:
+        Dictionary with status information
+    """
+    response = requests.post(f"{API_BASE_URL}/announcement", params={"text": text, "group_id": group_id})
+    response.raise_for_status()
+    return response.json()
+
+def get_group_state(group_id: int) -> Tuple[str, float]:
+    """
+    Get the status of the current announcement for a specific group.
+    
+    Args:
+        group_id: The ID of the group to check the announcement status for
+        
+    Returns:
+        Tuple containing (state_name, state_value)
+    """
+    response = requests.get(f"{API_BASE_URL}/groups/{group_id}/state")
+    response.raise_for_status()
+    state_data = response.json()
+    return (state_data["name"], state_data["value"])
+
+
+def get_available_endpoints() -> List[Endpoint]:
+    """Get all available endpoints."""
+    response = requests.get(f"{API_BASE_URL}/endpoints")
+    response.raise_for_status()
+    endpoints_dict = response.json()
+    # API returns a dictionary with endpoint IDs as keys
+    # Convert this to a list of Endpoint objects
+    return [Endpoint.model_validate(endpoint_data) for endpoint_id, endpoint_data in endpoints_dict.items()]
+
+def get_available_languages() -> List[str]:
+    """Get all available languages for announcements."""
+    response = requests.get(f"{API_BASE_URL}/languages")
+    response.raise_for_status()
+    return response.json()

src/multilang_translator/translator_config.py

@@ -0,0 +1,20 @@
+import os
+from pydantic import BaseModel
+
+VENV_DIR = os.path.join(os.path.dirname(__file__), './../../venv')
+
+class TranslatorLangConfig(BaseModel):
+    translator_llm: str = 'llama3.2:3b-instruct-q4_0' # TODO: this was migrated to translator_models - remove this
+    llm_client: str = 'ollama'
+    llm_host_url: str | None = 'http://localhost:11434'
+    llm_host_token: str | None = None
+    tts_system: str = 'piper'
+    tts_model: str ='de_DE-kerstin-low'
+
+class TranslatorConfig(BaseModel):
+    deu: TranslatorLangConfig = TranslatorLangConfig(tts_model = 'de_DE-thorsten-high')
+    eng: TranslatorLangConfig = TranslatorLangConfig(tts_model = 'en_GB-alba-medium')
+    fra: TranslatorLangConfig = TranslatorLangConfig(tts_model = 'fr_FR-siwis-medium')
+    spa: TranslatorLangConfig = TranslatorLangConfig(tts_model = 'es_ES-sharvard-medium')
+    ita: TranslatorLangConfig = TranslatorLangConfig(tts_model = 'it_IT-paola-medium')
+

src/multilang_translator/translator_models/__init__.py

@@ -0,0 +1 @@
+# Empty file to make the directory a package

src/multilang_translator/translator_models/translator_models.py

@@ -0,0 +1,58 @@
+"""
+Models for the translator API.
+Similar to the models used in auracaster-webui but simplified for the translator middleware.
+"""
+from enum import Enum
+from typing import List, Optional
+from pydantic import BaseModel
+
+
+class AnnouncementStates(Enum):
+    IDLE = 0
+    INIT = 0.1
+    TRANSLATING = 0.2
+    GENERATING_VOICE = 0.4
+    ROUTING = 0.6
+    BROADCASTING = 0.8
+    COMPLETED = 1
+    ERROR = 0
+
+
+class Endpoint(BaseModel):
+    """Defines an endpoint with its URL and capabilities."""
+    id: int
+    name: str
+    url: str
+    max_broadcasts: int = 1  # Maximum number of simultaneous broadcasts
+
+class TranslatorLangConfig(BaseModel):
+    translator_llm: str = 'llama3.2:3b-instruct-q4_0'
+    
+    llm_client: str = 'openwebui'  # remote (homserver)
+    llm_host_url: str = 'https://ollama.pstruebi.xyz'
+    llm_host_token: str = 'sk-17124cb84df14cc6ab2d9e17d0724d13'
+    # llm_client: str = 'ollama' #local
+    # llm_host_url: str | None = 'http://localhost:11434'
+    # llm_host_token: str | None = None
+    
+    tts_system: str = 'piper'
+    tts_model: str ='de_DE-kerstin-low'
+
+
+class TranslatorConfig(BaseModel):
+    deu: TranslatorLangConfig = TranslatorLangConfig(tts_model = 'de_DE-thorsten-high')
+    eng: TranslatorLangConfig = TranslatorLangConfig(tts_model = 'en_GB-alba-medium')
+    fra: TranslatorLangConfig = TranslatorLangConfig(tts_model = 'fr_FR-siwis-medium')
+    spa: TranslatorLangConfig = TranslatorLangConfig(tts_model = 'es_ES-sharvard-medium')
+    ita: TranslatorLangConfig = TranslatorLangConfig(tts_model = 'it_IT-paola-medium')
+
+
+class EndpointGroup(BaseModel):
+    id: int
+    name: str
+    languages: List[str]
+    endpoints: List[Endpoint]
+    sampling_rate_hz: int = 16000
+    translator_config: TranslatorConfig = TranslatorConfig()
+    current_state: AnnouncementStates = AnnouncementStates.IDLE
+    anouncement_start_time: float = 0.0

src/multilang_translator/translator_server/__init__.py

@@ -0,0 +1 @@
+# Empty file to make the directory a package

src/multilang_translator/translator_server/endpoints_db.py

@@ -0,0 +1,143 @@
+"""
+Database file for endpoint definitions.
+This file contains configurations for auracast endpoints including their IP addresses and capabilities.
+"""
+from typing import List, Optional
+from multilang_translator.translator_models.translator_models import EndpointGroup, Endpoint
+
+
+SUPPORTED_LANGUAGES = ["deu", "eng", "fra", "spa", "ita"]
+
+# Database of endpoints
+ENDPOINTS: dict[int: Endpoint] = { # for now make sure, .id and key are the same
+    0: Endpoint(
+        id=0,
+        name="Local Endpoint",
+        url="http://localhost:5000",
+        max_broadcasts=3,
+    ),
+    1: Endpoint(
+        id=1,
+        name="Gate 1",
+        url="http://pi3:5000",
+        max_broadcasts=3,
+    ),
+    2: Endpoint(
+        id=2,
+        name="Gate 2",
+        url="http://192.168.1.102:5000",
+        max_broadcasts=3,
+    ),
+}
+
+# Database of endpoint groups with default endpoints
+ENDPOINT_GROUPS: dict[int:EndpointGroup] = { # for now make sure , .id and key are the same
+    0: EndpointGroup(
+        id=0,
+        name="Local Group",
+        languages=["deu", "eng"],
+        endpoints=[ENDPOINTS[0]],
+    ),
+    1: EndpointGroup(
+        id=1,
+        name="Gate1",
+        languages=["deu", "fra"],
+        endpoints=[ENDPOINTS[1]],
+    )
+}
+
+def get_available_languages() -> List[str]:
+    """Get a list of all supported languages."""
+    return SUPPORTED_LANGUAGES
+
+# Endpoint functions
+def get_all_endpoints() -> List[Endpoint]:
+    """Get all active endpoints."""
+    return ENDPOINTS
+
+def get_endpoint_by_id(endpoint_id: str) -> Optional[Endpoint]:
+    """Get an endpoint by its ID."""
+    return ENDPOINTS[endpoint_id]
+
+
+def add_endpoint(endpoint: Endpoint) -> Endpoint:
+    """Add a new endpoint to the database."""
+    if endpoint.id in ENDPOINTS:
+        raise ValueError(f"Endpoint with ID {endpoint.id} already exists")
+    ENDPOINTS[endpoint.id] = endpoint
+    return endpoint
+
+
+def update_endpoint(endpoint_id: str, updated_endpoint: Endpoint) -> Endpoint:
+    """Update an existing endpoint in the database."""
+    if endpoint_id not in ENDPOINTS:
+        raise ValueError(f"Endpoint {endpoint_id} not found")
+    
+    # Ensure the ID is preserved
+    updated_endpoint.id = endpoint_id
+    ENDPOINTS[endpoint_id] = updated_endpoint
+    return updated_endpoint
+
+
+def delete_endpoint(endpoint_id: str) -> None:
+    """Delete an endpoint from the database."""
+    if endpoint_id not in ENDPOINTS:
+        raise ValueError(f"Endpoint {endpoint_id} not found")
+    
+    # Check if this endpoint is used in any groups
+    for group in ENDPOINT_GROUPS.values():
+        if endpoint_id in group.endpoints:
+            raise ValueError(f"Cannot delete endpoint {endpoint_id}, it is used in group {group.id}")
+    
+    del ENDPOINTS[endpoint_id]
+
+
+# Endpoint Group functions
+def get_all_groups() -> List[EndpointGroup]:
+    """Get all endpoint groups."""
+    return list(ENDPOINT_GROUPS.values())
+
+
+def get_group_by_id(group_id: int) -> Optional[EndpointGroup]:
+    """Get an endpoint group by its ID."""
+    return ENDPOINT_GROUPS.get(group_id)
+
+
+def add_group(group: EndpointGroup) -> EndpointGroup:
+    """Add a new endpoint group to the database."""
+    if group.id in ENDPOINT_GROUPS:
+        raise ValueError(f"Group with ID {group.id} already exists")
+    
+    # Validate that all referenced endpoints exist
+    for endpoint_id in group.endpoints:
+        if endpoint_id not in ENDPOINTS:
+            raise ValueError(f"Endpoint {endpoint_id} not found")
+    
+    ENDPOINT_GROUPS[group.id] = group
+    return group
+
+
+def update_group(group_id: int, updated_group: EndpointGroup) -> EndpointGroup:
+    """Update an existing endpoint group in the database."""
+    if group_id not in ENDPOINT_GROUPS:
+        raise ValueError(f"Group {group_id} not found")
+
+    # Validate that all referenced endpoints exist
+    for endpoint in updated_group.endpoints:
+        if endpoint.id not in ENDPOINTS.keys():
+            raise ValueError(f"Endpoint with id {endpoint.id} not found")
+
+    # Ensure the ID is preserved
+    updated_group.id = group_id
+    ENDPOINT_GROUPS[group_id] = updated_group
+    return updated_group
+
+
+def delete_group(group_id: int) -> None:
+    """Delete an endpoint group from the database."""
+    if group_id not in ENDPOINT_GROUPS:
+        raise ValueError(f"Group {group_id} not found")
+    
+    del ENDPOINT_GROUPS[group_id]
+
+

src/multilang_translator/translator_server/main_translator_api.py

@@ -0,0 +1,28 @@
+"""
+Entry point for the Translator API server.
+This file starts the FastAPI server with the translator_server.
+"""
+import uvicorn
+import logging as log
+import sys
+import os
+
+# Add the parent directory to the Python path to find the multilang_translator package
+current_dir = os.path.dirname(os.path.abspath(__file__))
+parent_dir = os.path.dirname(os.path.dirname(os.path.dirname(current_dir)))
+if parent_dir not in sys.path:
+    sys.path.insert(0, parent_dir)
+
+if __name__ == "__main__":
+    log.basicConfig(
+        level=log.INFO,
+        format='%(module)s.py:%(lineno)d %(levelname)s: %(message)s'
+    )
+    log.info("Starting Translator API server")
+    uvicorn.run(
+        "multilang_translator.translator_server.translator_server:app",
+        host="0.0.0.0",
+        port=7999,
+        reload=True,
+        log_level="debug"
+    )

src/multilang_translator/translator_server/translator_server.py

@@ -0,0 +1,346 @@
+"""
+FastAPI implementation of the Multilang Translator API.
+This API mimics the mock_api from auracaster-webui to allow integration.
+"""
+import time
+import logging as log
+import asyncio
+import random
+
+from fastapi import FastAPI, HTTPException
+from fastapi.middleware.cors import CORSMiddleware
+
+# Import models
+from multilang_translator.translator_models.translator_models import AnnouncementStates, Endpoint, EndpointGroup
+from multilang_translator.translator import llm_translator
+from multilang_translator.translator_server import endpoints_db
+from voice_provider import text_to_speech
+
+# Import the endpoints database and multicast client
+from auracast import multicast_client, auracast_config
+
+# Create FastAPI app
+app = FastAPI()
+
+# Add CORS middleware to allow cross-origin requests
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+# Endpoint configuration cache
+CURRENT_ENDPOINT_CONFIG = {}
+
+async def init_endpoint(endpoint: Endpoint, languages: list[str], sampling_rate_hz: int):
+    """Initialize a specific endpoint for multicast."""
+
+    current_config = CURRENT_ENDPOINT_CONFIG.get(endpoint.id)
+
+    if current_config is not None:
+        current_langs = [big.language for big in current_config.bigs]
+        # if languages are unchanged and the caster client status is initiailized, skip init
+        if current_langs == languages:
+            # Get status asynchronously
+            status = await multicast_client.get_status(base_url=endpoint.url)
+            if status['is_initialized']:
+                log.info('Endpoint %s was already initialized', endpoint.name)
+                return
+
+    log.info(f"Initializing endpoint: {endpoint.name} at {endpoint.url}")
+    # Load a default config
+    config = auracast_config.AuracastConfigGroup(
+        bigs=[getattr(auracast_config, f"AuracastBigConfig{lang.capitalize()}")()
+              for lang in languages]
+    )
+
+    # overwrite some default configs
+    config.transport = 'auto'
+    config.auracast_device_address = ':'.join(f"{random.randint(0, 255):02X}" for _ in range(6))
+    config.auracast_sampling_rate_hz = sampling_rate_hz
+
+    # Configure the bigs
+    for big in config.bigs:
+        big.loop = False
+        big.name = endpoint.name
+        big.random_address = ':'.join(f"{random.randint(0, 255):02X}" for _ in range(6))
+        big.id = random.randint(0, 2**16) #TODO: how many bits is this ?
+        #big.program_info = big.program_info + ' ' + endpoint.name
+
+    # make async init request
+    ret = await multicast_client.init(config, base_url=endpoint.url)
+    # if ret != 200: # TODO: this is not working, should probably be handled async
+    #     log.error('Init of endpoint %s was unsucessfull', endpoint.name)
+    #     raise Exception(f"Init was of endpoint {endpoint.name} was unsucessfull")
+    CURRENT_ENDPOINT_CONFIG[endpoint.id] = config.model_copy()
+    log.info(f"Endpoint {endpoint.name} initialized successfully")
+   #else:
+     #   log.info('Endpoint %s was already initialized', endpoint.name)
+
+
+async def make_announcement(text: str, ep_group: EndpointGroup):
+    """
+    Make an announcement to a group of endpoints.
+    """
+
+    if text == "":
+        log.warning("Announcement text is empty")
+        return {"error": "Announcement text is empty"}
+
+    ep_group.current_state = AnnouncementStates.IDLE
+    ep_group.anouncement_start_time = time.time()
+    # update the database with the new state and start time so this can be read by another process
+    endpoints_db.update_group(ep_group.id, ep_group)
+
+    # Initialize all endpoints in the group concurrently
+    ep_group.current_state = AnnouncementStates.INIT
+    endpoints_db.update_group(ep_group.id, ep_group)
+    
+    # Create init tasks and run them concurrently
+    init_tasks = [
+        init_endpoint(endpoint, ep_group.languages, ep_group.sampling_rate_hz)
+        for endpoint in ep_group.endpoints
+    ]
+
+    # make sure init finished
+    await asyncio.gather(*init_tasks)
+
+    # Translate the text for each language (concurrently)
+    base_lang = "deu"  # German is the base language
+    target_langs = ep_group.languages.copy()
+    if base_lang in target_langs:
+        target_langs.remove(base_lang)
+
+    ep_group.current_state = AnnouncementStates.TRANSLATING
+    endpoints_db.update_group(ep_group.id, ep_group)
+
+    # Create translation tasks
+    translations = {base_lang: text}
+    translation_tasks = []
+    
+    for lang in target_langs:
+            # Prepare translation task
+            trans_conf = getattr(ep_group.translator_config, lang)
+            task = llm_translator.translate_de_to_x_async(
+                text=text,
+                target_language=lang,
+                client=trans_conf.llm_client,
+                model=trans_conf.translator_llm,
+                host=trans_conf.llm_host_url,
+                token=trans_conf.llm_host_token
+            )
+            translation_tasks.append(task)
+    
+    # Wait for all translations to complete concurrently
+    results = await asyncio.gather(*translation_tasks)
+    for i, translation in enumerate(results):
+        lang = target_langs[i]
+        translations[lang] = translation
+        log.info(f"Translated to {lang}: {translation}")
+
+    # Generate voices concurrently
+    ep_group.current_state = AnnouncementStates.GENERATING_VOICE
+    endpoints_db.update_group(ep_group.id, ep_group)
+
+    # Prepare synthesis tasks and run them concurrently
+    synth_langs = ep_group.languages
+    synthesis_tasks = []
+    for lang in synth_langs:
+        trans_conf = getattr(ep_group.translator_config, lang)
+        task = text_to_speech.synthesize_async(
+            translations[lang],
+            ep_group.sampling_rate_hz,
+            trans_conf.tts_system,
+            trans_conf.tts_model,
+            return_lc3=True
+        )
+        synthesis_tasks.append(task)
+    
+    # Wait for all synthesis tasks to complete concurrently
+    audio = {}
+    if synthesis_tasks:
+        results = await asyncio.gather(*synthesis_tasks)
+        for i, audio_data in enumerate(results):
+            audio[synth_langs[i]] = audio_data
+
+    # Start the monitoring coroutine to wait for streaming to complete
+    # This will set the state to COMPLETED when finished
+    asyncio.create_task(monitor_streaming_completion(ep_group))
+
+    # Broadcast to all endpoints in group concurrently
+    broadcast_tasks = []
+    for endpoint in ep_group.endpoints:
+        log.info(f"Broadcasting to {endpoint.name} for languages: {', '.join(audio.keys())}")
+        task = multicast_client.send_audio(audio, base_url=endpoint.url)
+        broadcast_tasks.append(task)
+    
+    # Wait for all broadcasts to complete
+    await asyncio.gather(*broadcast_tasks)
+
+    # Return the translations
+    return {"translations": translations}
+
+
+async def monitor_streaming_completion(ep_group: EndpointGroup):
+    """
+    Monitor streaming status after audio is sent and update group state when complete.
+
+    Args:
+        ep_group: The endpoint group being monitored
+    """
+    log.info(f"Starting streaming completion monitoring for endpoint group {ep_group.id}")
+
+
+    # Set a shorter timeout as requested
+    max_completion_time = 60  # seconds
+
+    # First check if we are actually in streaming state
+    streaming_started = False
+    initial_check_timeout = 10  # seconds
+    initial_check_start = time.time()
+
+    # Wait for streaming to start (with timeout)
+    while time.time() - initial_check_start < initial_check_timeout:
+        # Wait before checking again
+        await asyncio.sleep(1)
+
+        any_streaming = False
+        for endpoint in ep_group.endpoints:
+            status = await multicast_client.get_status(base_url=endpoint.url)
+            if status.get("is_streaming", False):
+                any_streaming = True
+                log.info(f"Streaming confirmed started on endpoint {endpoint.name}")
+                break
+
+        if any_streaming:
+            streaming_started = True
+            break
+
+    if not streaming_started:
+        log.warning(f"No endpoints started streaming for group {ep_group.id} after {initial_check_timeout}s")
+        # Still update to completed since there's nothing to wait for
+        ep_group.current_state = AnnouncementStates.ERROR
+        endpoints_db.update_group(ep_group.id, ep_group)
+        return
+
+    # Update group progress
+    ep_group.current_state = AnnouncementStates.BROADCASTING
+    endpoints_db.update_group(ep_group.id, ep_group)
+
+    # Now monitor until streaming completes on all endpoints
+    check_completion_start_time = time.time()
+    completed = [False for _ in ep_group.endpoints]
+    while not all(completed) or time.time() - check_completion_start_time > max_completion_time:
+        await asyncio.sleep(1)
+
+        # Check status of each endpoint
+        for i, endpoint in enumerate(ep_group.endpoints):
+            status = await multicast_client.get_status(base_url=endpoint.url)
+            completed[i] = not status['is_streaming']
+
+    if all(completed):
+        log.info(f"All endpoints completed streaming for group {ep_group.id}")
+        # Update group state to completed
+        ep_group.current_state = AnnouncementStates.COMPLETED
+        endpoints_db.update_group(ep_group.id, ep_group)
+        log.info(f"Updated group {ep_group.id} state to COMPLETED")
+
+    else:
+        log.error(f"Max wait time reached for group {ep_group.id}. Forcing completion.")
+
+
+@app.get("/groups")
+async def get_groups():
+    """Get all endpoint groups with their current status."""
+    return endpoints_db.get_all_groups()
+
+
+@app.post("/groups")
+async def create_group(group: endpoints_db.EndpointGroup):
+    """Add a new endpoint group."""
+    try:
+        return endpoints_db.add_group(group)
+    except ValueError as e:
+        raise HTTPException(status_code=400, detail=str(e))
+
+@app.get("/groups/{group_id}/state") # TODO: think about progress tracking
+async def get_group_state(group_id: int):
+    """Get the status of a specific endpoint."""
+    # Check if the endpoint exists
+    ep_group = endpoints_db.get_group_by_id(group_id)
+    if not ep_group:
+        raise HTTPException(status_code=404, detail=f"Endpoint {group_id} not found")
+
+    return {"name": ep_group.current_state.name, "value": ep_group.current_state.value}
+
+
+@app.put("/groups/{group_id}")
+async def update_group(group_id: int, updated_group: endpoints_db.EndpointGroup):
+    """Update an existing endpoint group."""
+    try:
+        return endpoints_db.update_group(group_id, updated_group)
+    except ValueError as e:
+        raise HTTPException(status_code=400, detail=str(e))
+
+
+@app.delete("/groups/{group_id}")
+async def delete_group(group_id: int):
+    """Delete an endpoint group."""
+    try:
+        endpoints_db.delete_group(group_id)
+        return {"message": f"Group {group_id} deleted successfully"}
+    except ValueError as e:
+        raise HTTPException(status_code=400, detail=str(e))
+
+
+@app.post("/announcement")
+async def start_announcement(text: str, group_id: int):
+    """Start a new announcement to the specified endpoint group."""
+    global announcement_task
+
+    # Get the group from active groups or database
+    group = endpoints_db.get_group_by_id(group_id)
+    if not group:
+        raise HTTPException(status_code=400, detail=f"Group {group_id} not found")
+
+    # Check if we're already processing an announcement
+    #if announcement_task and not announcement_task.done():
+    #    raise HTTPException(status_code=400, detail="Already processing an announcement")
+
+    # Start the announcement task
+    announcement_task = asyncio.create_task(make_announcement(text, group))
+    return {"status": "Announcement started", "group_id": group_id}
+
+
+
+@app.get("/endpoints")
+async def get_available_endpoints():
+    """Get all available endpoints with their capabilities."""
+    return endpoints_db.get_all_endpoints()
+
+
+@app.get("/languages")
+async def get_available_languages():
+    """Get all available languages for announcements."""
+    return endpoints_db.get_available_languages()
+
+
+
+if __name__ == "__main__":
+    import uvicorn
+    log.basicConfig(
+        level=log.DEBUG,
+        format='%(module)s.py:%(lineno)d %(levelname)s: %(message)s'
+    )
+    # with reload=True logging of modules does not function as expected
+    uvicorn.run(
+        app,
+        #'translator_server:app',
+        host="0.0.0.0",
+        port=7999,
+        #reload=True,
+        #log_config=None,
+        #log_level="info"
+        )

src/voice_client/tts_client.py

@@ -0,0 +1,44 @@
+import requests
+import numpy as np
+import soundfile as sf
+
+from voice_models.request_models import SynthesizeRequest
+
+
+API_URL = "http://127.0.0.1:8099/synthesize/"
+
+def request_synthesis(request_data: SynthesizeRequest):
+    response = requests.post(API_URL, json=request_data.model_dump())
+
+    if response.status_code == 200:
+        response_data = response.json()
+        
+        if request_data.return_lc3:
+            # Save LC3 audio as binary file
+            lc3_bytes = bytes.fromhex(response_data["audio_lc3"])
+            return lc3_bytes
+        
+        else:
+            # Convert hex-encoded PCM bytes back to numpy array and save as WAV
+            audio_bytes = bytes.fromhex(response_data["audio_pcm"])
+            audio_array = np.frombuffer(audio_bytes, dtype=np.float32)
+            return audio_array
+    
+    else:
+        print(f"Error: {response.status_code}, {response.text}")
+
+if __name__ == "__main__":
+
+    target_rate=16000
+
+    # Example request
+    request_data = SynthesizeRequest(
+        text="Hello, this is a test.",
+        target_sample_rate=target_rate,
+        framework="piper",
+        model="de_DE-kerstin-low",
+        return_lc3=False  # Set to True to receive LC3 compressed output
+    )
+
+    audio = request_synthesis(request_data)
+    sf.write('hello.wav', audio, target_rate)

src/voice_models/request_models.py

@@ -0,0 +1,9 @@
+from pydantic import BaseModel
+
+class SynthesizeRequest(BaseModel):
+    text: str
+    target_sample_rate: int = 16000
+    framework: str = "piper"
+    model: str = "en_US-lessac-medium"
+    return_lc3: bool = False
+

src/voice_provider/text_to_speech.py

@@ -0,0 +1,123 @@
+import os
+import shutil
+import subprocess
+import time
+import json
+import logging as log
+import numpy as np
+import asyncio
+from voice_provider.utils.resample import resample_array
+from voice_provider.utils.encode_lc3 import encode_lc3
+
+PIPER_EXE = shutil.which('piper')
+
+TTS_DIR = os.path.join(os.path.dirname(__file__))
+PIPER_WORKDIR = f'{TTS_DIR}/piper'
+
+if not PIPER_EXE:
+    PIPER_EXE = f'{TTS_DIR}/../../venv/bin/piper'
+
+def synth_piper(text, model="en_US-lessac-medium"):
+    pwd = os.getcwd()
+    os.chdir(PIPER_WORKDIR)
+    start = time.time()
+    
+    # make sure piper has voices.json in working directory, otherwise it attempts to always load models
+    ret = subprocess.run( # TODO: wrap this whole thing in a class and open a permanent pipe to the model 
+        [
+            PIPER_EXE, 
+            '--cuda',
+            '--model', model, 
+            '--output-raw'
+        ], 
+        input=text.encode('utf-8'),
+        capture_output=True
+        )
+    os.chdir(pwd)
+
+    log.warning('Piper stderr:\n%s', ret.stderr)
+    assert ret.returncode == 0, 'Piper returncode was not 0.'
+    audio = ret.stdout
+
+    log.info("Running piper for model %s took %s s", model, round(time.time() - start, 3))
+
+    with open (f'{PIPER_WORKDIR}/{model}.onnx.json') as f: # TODO: wrap everyth0ing into a class, store the json permanently
+        model_json = json.load(f)
+
+    return model_json, audio
+
+
+def synthesize(
+        text, 
+        target_sample_rate, 
+        framework, 
+        model="en_US-lessac-medium", 
+        return_lc3=True
+        ):
+
+    if framework == 'piper':
+        model_json, audio_raw = synth_piper(text, model)
+        tts_sample_rate = model_json['audio']['sample_rate']
+        audio_np = np.frombuffer(audio_raw, dtype=np.dtype('<i2')).astype(np.float32) /(2**15-1)# convert to float fraction
+        audio = resample_array(audio_np, tts_sample_rate, target_sample_rate)
+        
+    elif framework == 'koro':
+        pass
+    elif framework == 'xtts':
+        pass
+    elif framework == 'zonos':
+        pass
+    else: raise NotImplementedError('unknown framework')
+
+    if return_lc3:
+        audio_pcm = (audio * 2**15-1).astype(np.int16)
+        lc3 = encode_lc3(audio_pcm, target_sample_rate, 40) # TODO: octetts per frame should be parameter
+        return lc3
+    else:
+        return audio
+
+
+async def synthesize_async(
+        text, 
+        target_sample_rate, 
+        framework, 
+        model="en_US-lessac-medium", 
+        return_lc3=True
+        ):
+    """
+    Asynchronous version of the synthesize function that runs in a thread pool.
+    
+    Args:
+        text: Text to synthesize
+        target_sample_rate: Target sample rate for the audio
+        framework: TTS framework to use (e.g., 'piper')
+        model: Model to use for synthesis
+        return_lc3: Whether to return LC3-encoded audio
+        
+    Returns:
+        LC3-encoded audio as string or raw audio as numpy array
+    """
+    # Run the CPU-intensive synthesis in a thread pool
+    loop = asyncio.get_event_loop()
+    result = await loop.run_in_executor(
+        None,
+        lambda: synthesize(text, target_sample_rate, framework, model, return_lc3)
+    )
+    return result
+
+
+if __name__ == '__main__':
+    import logging
+    import soundfile as sf
+
+    logging.basicConfig(
+        level=logging.INFO,
+        format='%(module)s.py:%(lineno)d %(levelname)s: %(message)s'
+    )
+    target_rate=16000
+
+    audio = synthesize('Hello World', target_rate, 'piper', model= 'de_DE-kerstin-low', return_lc3=False)
+
+    sf.write('hello.wav', audio, target_rate)
+
+    print('Done.')

src/voice_provider/tts_server.py

@@ -0,0 +1,43 @@
+from fastapi import FastAPI, HTTPException
+import numpy as np
+
+from voice_models.request_models import SynthesizeRequest
+from voice_provider.text_to_speech import synthesize_async
+
+app = FastAPI()
+
+HOST_PORT = 8099
+
+
+@app.post("/synthesize/")
+async def synthesize_speech(request: SynthesizeRequest):
+    try:
+        audio = await synthesize_async(
+            text=request.text,
+            target_sample_rate=request.target_sample_rate,
+            framework=request.framework,
+            model=request.model,
+            return_lc3=request.return_lc3
+        )
+
+        if request.return_lc3:
+            # If it's already a string (LC3 data), convert it to bytes for hex encoding
+            if isinstance(audio, str):
+                audio_bytes = audio.encode('latin-1')
+                return {"audio_lc3": audio_bytes.hex()}
+            # If it's already bytes
+            elif isinstance(audio, bytes):
+                return {"audio_lc3": audio.hex()}
+            else:
+                raise ValueError(f"Unexpected audio type: {type(audio)}")
+        else:
+            # If it's numpy array (non-LC3), convert to bytes
+            audio_bytes = audio.astype(np.float32).tobytes()
+            return {"audio_pcm": audio_bytes.hex()}
+        
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+if __name__ == "__main__":
+    import uvicorn
+    uvicorn.run(app, host="127.0.0.1", port=HOST_PORT)

src/voice_provider/utils/encode_lc3.py

@@ -0,0 +1,32 @@
+import numpy as np
+import lc3
+
+def encode_lc3(
+        audio: np.array,
+        output_sample_rate_hz,
+        octets_per_frame,
+        frame_duration_us=10000,
+        pcm_bit_depth = 16
+        ):
+
+    encoder = lc3.Encoder(
+        frame_duration_us=frame_duration_us,
+        sample_rate_hz=output_sample_rate_hz,
+        num_channels=1,
+        #input_sample_rate_hz=input_sample_rate,
+    )
+
+    lc3_frame_samples = encoder.get_frame_samples() # number of the pcm samples per lc3 frame
+
+    # reshape array into slices of lc3_frame_samples and padd with zeros
+    pad_width = (lc3_frame_samples - len(audio) % lc3_frame_samples) % lc3_frame_samples  # Compute padding length
+    arr_padded = np.pad(audio, (0, pad_width), mode='constant', constant_values=0)
+    reshaped_arr = arr_padded.reshape(-1, lc3_frame_samples)
+
+    lc3_bytes = b''
+    for pcm_frame in reshaped_arr:
+        lc3_bytes += encoder.encode(
+                pcm_frame, num_bytes=octets_per_frame, bit_depth=pcm_bit_depth
+            )
+
+    return lc3_bytes

src/voice_provider/utils/resample.py

@@ -0,0 +1,48 @@
+# resample .wav source to target sampling rate
+import logging as log
+import time
+import os
+import librosa
+import soundfile as sf
+
+
+def resample_file(filename, out_filename, target_rate):
+    start=time.time()
+    # Load the original audio file
+    audio, rate = librosa.load(filename)
+
+    if rate == target_rate: # Nothing to do 
+        sf.write(out_filename, audio, target_rate)
+        return
+
+    # Convert the sample rate to 24 kHz
+    resampled_audio = librosa.resample(audio, orig_sr=rate, target_sr=target_rate)
+
+    # Save the resampled audio as a new .wav file
+    sf.write(out_filename, resampled_audio, target_rate)
+    
+    log.info("Resampling of %s took  %s s", os.path.basename(filename), round(time.time() - start, 3))
+
+
+def resample_array(audio, rate, target_rate):
+    start=time.time()
+    # Load the original audio file
+
+    if rate == target_rate: # Nothing to do 
+        log.info('audio already at target rate, skipping resample')
+        return audio
+
+    # Convert the sample rate to target rate
+    resampled_audio = librosa.resample(audio, orig_sr=rate, target_sr=target_rate)
+
+    # Save the resampled audio as a new .wav file
+    
+    log.info("Resampling took  %s s", round(time.time() - start, 3))
+    return resampled_audio
+
+
+if __name__ == "__main__":
+    import os
+    os.chdir(os.path.dirname(__file__))
+    file_dir = '../text_to_speech/'
+    resample_file(f'{file_dir}/welcome.wav', f'{file_dir}/welcome_resampled.wav')

tests/get_group_state.py

@@ -0,0 +1,9 @@
+import requests
+import time
+
+if __name__ == '__main__':
+    # get the group state every 0.5s 
+    while True:
+        response = requests.get('http://localhost:7999/groups/0/state')
+        print(response.json())
+        time.sleep(0.5)

tests/test_backend.py

@@ -4,7 +4,7 @@ import time
 import os
 import subprocess
 
-from multilang_translator.config import LANG_CONFIG
+from multilang_translator.translator_config import LANG_CONFIG
 from multilang_translator.backend_controller.broadcaster_play_once import broadcaster_play_file
 from multilang_translator.backend_controller.broadcaster_copy_files import copy_to_broadcaster
 

tests/test_system.py

@@ -1,4 +1,4 @@
-from multilang_translator.main import announcement_from_german_text
+from multilang_translator.main_local import announcement_from_german_text
 from multilang_translator.translator import test_content
 
 

tests/test_translator.py

@@ -1,6 +1,6 @@
 from multilang_translator.translator.llm_translator import translator_de_en, translator_de_fr, translator_de_it
 from multilang_translator.translator.test_content import TESTSENTENCE_DE_BROKER, TESTSENTENCE_DE_RAINBOW
-from multilang_translator.main import translate_from_german
+from multilang_translator.main_local import translate_from_german
 
 
 import time