Azure AI services - Speech service

 Azure AI services - Speech service

Source: https://github.com/MicrosoftLearning/mslearn-ai-language

1. Create 'Speech service' in Azure - copy key and region

C# Code:

dotnet add package Microsoft.CognitiveServices.Speech --version 1.30.0

dotnet add package System.Windows.Extensions --version 4.6.0 

using System;

using System.Threading.Tasks;

using Microsoft.Extensions.Configuration;

using System.Collections.Generic;

using System.Text;

// Import namespaces

using Microsoft.CognitiveServices.Speech;

using Microsoft.CognitiveServices.Speech.Audio;

using Microsoft.CognitiveServices.Speech.Translation;

using System.Media;

namespace speech_translation

{

    class Program

    {

        private static SpeechConfig speechConfig;

        private static SpeechTranslationConfig translationConfig;

        static async Task Main(string[] args)

        {

            try

            {

                // Get config settings from AppSettings

                IConfigurationBuilder builder =                 new ConfigurationBuilder().AddJsonFile("appsettings.json");

                IConfigurationRoot configuration = builder.Build();

                string aiSvcKey = "1RBACOGNMJb"; //configuration["SpeechKey"];

                string aiSvcRegion = "eastus"; //configuration["SpeechRegion"];

                // Set console encoding to unicode

                Console.InputEncoding = Encoding.Unicode;

                Console.OutputEncoding = Encoding.Unicode;

                // Configure translation

                translationConfig = SpeechTranslationConfig.FromSubscription(aiSvcKey,                 aiSvcRegion);

                translationConfig.SpeechRecognitionLanguage = "en-US";

                translationConfig.AddTargetLanguage("fr");

                translationConfig.AddTargetLanguage("es");

                translationConfig.AddTargetLanguage("hi");

                Console.WriteLine("Ready to translate from " +                 translationConfig.SpeechRecognitionLanguage);

                // Configure speech

                speechConfig = SpeechConfig.FromSubscription(aiSvcKey, aiSvcRegion);

                string targetLanguage = "";

                while (targetLanguage != "quit")

                {

                    Console.WriteLine("\nEnter a target language\n fr = French\n es =                     Spanish\n hi = Hindi\n Enter anything else to stop\n");

                    targetLanguage = Console.ReadLine().ToLower();

                    if (translationConfig.TargetLanguages.Contains(targetLanguage))

                    {

                        await Translate(targetLanguage);

                    }

                    else

                    {

                        targetLanguage = "quit";

                    }

                }

            }

            catch (Exception ex)

            {

                Console.WriteLine(ex.Message);

            }

        }

        static async Task Translate(string targetLanguage)

        {

            string translation = "";

            // Translate speech

            using AudioConfig audioConfig = AudioConfig.FromDefaultMicrophoneInput();

            using TranslationRecognizer translator = new TranslationRecognizer(translationConfig,             audioConfig);

            Console.WriteLine("Speak now...");

            TranslationRecognitionResult result = await translator.RecognizeOnceAsync();

            Console.WriteLine($"Translating '{result.Text}'");

            translation = result.Translations[targetLanguage];

            Console.OutputEncoding = Encoding.UTF8;

            Console.WriteLine(translation);

            // Translate speech

            // string audioFile = "station.wav";

            // SoundPlayer wavPlayer = new SoundPlayer(audioFile);

            // wavPlayer.Play();

            // using AudioConfig audioConfig = AudioConfig.FromWavFileInput(audioFile);

            // using TranslationRecognizer translator = new TranslationRecognizer(translationConfig,             audioConfig);

            // Console.WriteLine("Getting speech from file...");

            // TranslationRecognitionResult result = await translator.RecognizeOnceAsync();

            // Console.WriteLine($"Translating '{result.Text}'");

            // translation = result.Translations[targetLanguage];

            // Console.OutputEncoding = Encoding.UTF8;

            // Console.WriteLine(translation);

            // Synthesize translation

            var voices = new Dictionary<string, string>

            {

                ["fr"] = "fr-FR-HenriNeural",

                ["es"] = "es-ES-ElviraNeural",

                ["hi"] = "hi-IN-MadhurNeural"

            };

            speechConfig.SpeechSynthesisVoiceName = voices[targetLanguage];

            using SpeechSynthesizer speechSynthesizer = new SpeechSynthesizer(speechConfig);

            SpeechSynthesisResult speak = await speechSynthesizer.SpeakTextAsync(translation);

            if (speak.Reason != ResultReason.SynthesizingAudioCompleted)

            {

                Console.WriteLine(speak.Reason);

            }

        }

    }

}

OutPut:

Python Code:

pip install azure-cognitiveservices-speech==1.30.0

pip install playsound==1.3.0

from dotenv import load_dotenv

from datetime import datetime

import os

# Import namespaces

import azure.cognitiveservices.speech as speech_sdk

def main():

    try:

        global speech_config

        global translation_config

        # Get Configuration Settings

        load_dotenv()

        ai_key = '1RBACOGNMJb' #os.getenv('SPEECH_KEY')

        ai_region = 'eastus' # os.getenv('SPEECH_REGION')

        # Configure translation

        translation_config = speech_sdk.translation.SpeechTranslationConfig(subscription=ai_key,         region=ai_region)

        translation_config.speech_recognition_language = 'en-US'

        translation_config.add_target_language('fr')

        translation_config.add_target_language('es')

        translation_config.add_target_language('hi')

        print('Ready to translate from', translation_config.speech_recognition_language)

        # Configure speech

        speech_config = speech_sdk.SpeechConfig(subscription=ai_key, region=ai_region)

        # Get user input

        targetLanguage = ''

        while targetLanguage != 'quit':

            targetLanguage = input('\nEnter a target language\n fr = French\n es = Spanish\n hi =             Hindi\n Enter anything else to stop\n').lower()

            if targetLanguage in translation_config.target_languages:

                Translate(targetLanguage)

            else:

                targetLanguage = 'quit'

               

    except Exception as ex:

        print(ex)

def Translate(targetLanguage):

    translation = ''

    # Translate speech

    audio_config = speech_sdk.AudioConfig(use_default_microphone=True)

    translator = speech_sdk.translation.TranslationRecognizer(translation_config=translation_config,     audio_config=audio_config)

    print("Speak now...")

    result = translator.recognize_once_async().get()

    if result.reason == speech_sdk.ResultReason.TranslatedSpeech:

        print('Translating "{}"'.format(result.text))

    for language, translation in result.translations.items():

        print('Translation in {}: {}'.format(language, translation))

    else:

        print("No speech could be recognized or translation failed.")

    # Synthesize translation

    voices = {

        "fr": "fr-FR-HenriNeural",

        "es": "es-ES-ElviraNeural",

        "hi": "hi-IN-MadhurNeural"

    }

 

    # Assuming `targetLanguage` and `translation` are defined

    targetLanguage = "fr"  # Replace this with the actual target language

    translation = "Bonjour tout le monde"  # Replace this with the actual translation

 

    # Set the speech synthesis voice name based on the target language

    speech_config.speech_synthesis_voice_name = voices.get(targetLanguage)

 

    # Initialize the speech synthesizer

    speech_synthesizer = speech_sdk.SpeechSynthesizer(speech_config)

 

    # Synthesize the translated text to speech

    speak = speech_synthesizer.speak_text_async(translation).get()

 

    # Check if the synthesis was successful

    if speak.reason != speech_sdk.ResultReason.SynthesizingAudioCompleted:

        print(speak.reason)

    else:

        print("Speech synthesis completed successfully.")

if __name__ == "__main__":

    main()

OutPut: