Text to Speech, Speech to Text, and Speech Translation using Azure Cognetive Services

Azure Speech Services.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "!pip install azure-cognitiveservices-speech"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "ISfrwmurxjpj"
      },
      "outputs": [],
      "source": [
        "import azure.cognitiveservices.speech as speechsdk"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "FICAbruKyNsa"
      },
      "outputs": [],
      "source": [
        "SPEECH_KEY = \"<YOUR_API_KEY>\"\n",
        "SPEECH_REGION = \"<YOUR_REGION>\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "iFpblA213Dpj"
      },
      "outputs": [],
      "source": [
        "def text_to_speech(text: str, audio_config):\n",
        "  speech_config = speechsdk.SpeechConfig(subscription=SPEECH_KEY, region=SPEECH_REGION)\n",
        "\n",
        "  # The language of the voice that speaks\n",
        "  speech_config.speech_synthesis_voice_name = 'en-US-GuyNeural'\n",
        "\n",
        "  speech_synthesizer = speechsdk.SpeechSynthesizer(speech_config=speech_config, audio_config=audio_config)\n",
        "\n",
        "  speech_synthesis_result = speech_synthesizer.speak_text_async(text).get()\n",
        "\n",
        "  if speech_synthesis_result.reason == speechsdk.ResultReason.SynthesizingAudioCompleted:\n",
        "    print(\"Speech synthesized for text [{}]\".format(text))\n",
        "  elif speech_synthesis_result.reason == speechsdk.ResultReason.Canceled:\n",
        "      cancellation_details = speech_synthesis_result.cancellation_details\n",
        "      print(\"Speech synthesis canceled: {}\".format(cancellation_details.reason))\n",
        "      if cancellation_details.reason == speechsdk.CancellationReason.Error:\n",
        "          if cancellation_details.error_details:\n",
        "              print(\"Error details: {}\".format(cancellation_details.error_details))\n",
        "              print(\"Did you set the speech resource key and region values?\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "E5o2JknezCVB"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Speech synthesized for text [New York is a great city to visit]\n"
          ]
        }
      ],
      "source": [
        "text = \"New York is a great city to visit\"\n",
        "audio_config = speechsdk.audio.AudioOutputConfig(filename=\"file1.mp3\")\n",
        "text_to_speech(text, audio_config)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "id": "U9a9cHXs4nAq"
      },
      "outputs": [],
      "source": [
        "def speech_to_text():\n",
        "    speech_config = speechsdk.SpeechConfig(subscription=SPEECH_KEY, region=SPEECH_REGION)\n",
        "    speech_config.speech_recognition_language = \"en-US\"\n",
        "\n",
        "    audio_config = speechsdk.audio.AudioConfig(use_default_microphone=True)\n",
        "    speech_recognizer = speechsdk.SpeechRecognizer(speech_config=speech_config, audio_config=audio_config)\n",
        "\n",
        "    print(\"Speak into your microphone.\")\n",
        "    speech_recognition_result = speech_recognizer.recognize_once_async().get()\n",
        "\n",
        "    if speech_recognition_result.reason == speechsdk.ResultReason.RecognizedSpeech:\n",
        "        print(\"Recognized: {}\".format(speech_recognition_result.text))\n",
        "    elif speech_recognition_result.reason == speechsdk.ResultReason.NoMatch:\n",
        "        print(\"No speech could be recognized: {}\".format(speech_recognition_result.no_match_details))\n",
        "    elif speech_recognition_result.reason == speechsdk.ResultReason.Canceled:\n",
        "        cancellation_details = speech_recognition_result.cancellation_details\n",
        "        print(\"Speech Recognition canceled: {}\".format(cancellation_details.reason))\n",
        "        if cancellation_details.reason == speechsdk.CancellationReason.Error:\n",
        "            print(\"Error details: {}\".format(cancellation_details.error_details))\n",
        "            print(\"Did you set the speech resource key and region values?\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "speech_to_text()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {},
      "outputs": [],
      "source": [
        "def translate_speech_to_text(target_language: str, audio_config):\n",
        "    speech_translation_config = speechsdk.translation.SpeechTranslationConfig(subscription=SPEECH_KEY, region=SPEECH_REGION)\n",
        "    speech_translation_config.speech_recognition_language = \"en-US\"\n",
        "\n",
        "    speech_translation_config.add_target_language(target_language)\n",
        "\n",
        "    translation_recognizer = speechsdk.translation.TranslationRecognizer(translation_config=speech_translation_config, audio_config=audio_config)\n",
        "\n",
        "    translation_recognition_result = translation_recognizer.recognize_once()\n",
        "\n",
        "    if translation_recognition_result.reason == speechsdk.ResultReason.TranslatedSpeech:\n",
        "        print(\"Recognized: {}\".format(translation_recognition_result.text))\n",
        "        print(\"\"\"Translated into '{}': {}\"\"\".format(\n",
        "            target_language, \n",
        "            translation_recognition_result.translations[target_language]))\n",
        "    elif translation_recognition_result.reason == speechsdk.ResultReason.NoMatch:\n",
        "        print(\"No speech could be recognized: {}\".format(translation_recognition_result.no_match_details))\n",
        "    elif translation_recognition_result.reason == speechsdk.ResultReason.Canceled:\n",
        "        cancellation_details = translation_recognition_result.cancellation_details\n",
        "        print(\"Speech Recognition canceled: {}\".format(cancellation_details.reason))\n",
        "        if cancellation_details.reason == speechsdk.CancellationReason.Error:\n",
        "            print(\"Error details: {}\".format(cancellation_details.error_details))\n",
        "            print(\"Did you set the speech resource key and region values?\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Recognized: New York is a great city to visit.\n",
            "Translated into 'pt': Nova York é uma ótima cidade para visitar.\n"
          ]
        }
      ],
      "source": [
        "audio_config = speechsdk.audio.AudioConfig(filename=\"file1.mp3\")\n",
        "translate_speech_to_text(\"pt\", audio_config)"
      ]
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.7.13"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}