# Copyright 2025 Google LLC
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     https://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

%%bash
# Detect the operating system
os=$(uname -s)

if [[ "$os" == "Linux" ]]; then
  # Linux installation
  sudo apt update -y -qq
  sudo apt install ffmpeg -y -qq
  echo "ffmpeg installed successfully on Linux."
elif [[ "$os" == "Darwin" ]]; then
  # macOS installation
  if command -v brew &> /dev/null; then
    brew install ffmpeg
    if [[ $? -eq 0 ]]; then
        echo "ffmpeg installed successfully on macOS using Homebrew."
    else
        echo "Error installing ffmpeg on macOS using Homebrew."
    fi
  else
    echo "Homebrew is not installed. Please install Homebrew and try again."
  fi
else
  echo "Unsupported operating system: $os"
fi

%pip install --upgrade --quiet google-cloud-texttospeech

import sys

if "google.colab" in sys.modules:
    from google.colab import auth

    auth.authenticate_user()

# Use the environment variable if the user doesn't provide Project ID.
import os

# fmt: off
PROJECT_ID = "[your-project-id]"  # @param {type: "string", placeholder: "[your-project-id]", isTemplate: true}
# fmt: on
if not PROJECT_ID or PROJECT_ID == "[your-project-id]":
    PROJECT_ID = str(os.environ.get("GOOGLE_CLOUD_PROJECT"))

TTS_LOCATION = "global"

! gcloud config set project {PROJECT_ID}
! gcloud auth application-default set-quota-project {PROJECT_ID}
! gcloud auth application-default login -q

import re
from collections.abc import Iterator

import numpy as np
from IPython.display import Audio, display
from google.api_core.client_options import ClientOptions
from google.cloud import texttospeech_v1beta1 as texttospeech

API_ENDPOINT = (
    f"{TTS_LOCATION}-texttospeech.googleapis.com"
    if TTS_LOCATION != "global"
    else "texttospeech.googleapis.com"
)

client = texttospeech.TextToSpeechClient(
    client_options=ClientOptions(api_endpoint=API_ENDPOINT)
)

def text_generator(text: str) -> Iterator[str]:
    """Split text into sentences to simulate streaming"""
    # Use regex with positive lookahead to find sentence boundaries
    # without consuming the space after the punctuation
    sentences: list[str] = re.findall(r"[^.!?]+[.!?](?:\s|$)", text + " ")

    # Yield each complete sentence
    for sentence in sentences:
        yield sentence.strip()

    # Check if there's remaining text not caught by the regex
    # (text without ending punctuation)
    last_char_pos: int = 0
    for sentence in sentences:
        last_char_pos += len(sentence)

    if last_char_pos < len(text.strip()):
        remaining: str = text.strip()[last_char_pos:]
        if remaining:
            yield remaining.strip()


def process_streaming_audio(
    text: str,
    voice: texttospeech.VoiceSelectionParams,
    display_individual_chunks: bool = False,
) -> np.ndarray:
    """Process text into speech using streaming TTS"""
    # Generate sentences from text
    sentences: list[str] = list(text_generator(text))

    # Get streaming audio
    print("Streaming audio processing...")
    audio_iterator: Iterator[bytes] = synthesize_streaming(iter(sentences), voice=voice)

    # Process audio chunks
    final_audio_data: np.ndarray = np.array([], dtype=np.int16)

    for idx, audio_content in enumerate(audio_iterator):
        audio_chunk: np.ndarray = np.frombuffer(audio_content, dtype=np.int16)

        # Concatenate to final audio
        final_audio_data = np.concatenate((final_audio_data, audio_chunk))

        # Optionally display individual chunks
        if display_individual_chunks and len(audio_chunk) > 0:
            print(f"Processed chunk # {idx}")
            display(Audio(audio_chunk, rate=24000))

    print("Streaming audio processing complete!")
    return final_audio_data


def synthesize_streaming(
    text_iterator: Iterator[str],
    voice: texttospeech.VoiceSelectionParams,
) -> Iterator[bytes]:
    """Synthesizes speech from an iterator of text inputs and yields audio content as an iterator.

    This function demonstrates how to use the Google Cloud Text-to-Speech API
    to synthesize speech from a stream of text inputs provided by an iterator.
    It yields the audio content from each response as an iterator of bytes.

    """
    config_request = texttospeech.StreamingSynthesizeRequest(
        streaming_config=texttospeech.StreamingSynthesizeConfig(
            voice=voice,
        )
    )

    def request_generator() -> Iterator[texttospeech.StreamingSynthesizeRequest]:
        yield config_request
        for text in text_iterator:
            yield texttospeech.StreamingSynthesizeRequest(
                input=texttospeech.StreamingSynthesisInput(text=text)
            )

    streaming_responses: Iterator[texttospeech.StreamingSynthesizeResponse] = (
        client.streaming_synthesize(request_generator())
    )

    for response in streaming_responses:
        yield response.audio_content

# fmt: off
prompt = "Hello world! I am Chirp 3"  # @param ["Hallo Welt! Ich bin Chirp 3", "Hello world! I am Chirp 3", "¡Hola mundo! Soy Chirp 3", "Bonjour le monde ! Je suis Chirp 3", "नमस्ते दुनिया! मैं चिर्प 3 हूं", "Olá Mundo! Eu sou o Chirp 3", "مرحبا بالعالم! أنا تشيرب 3", "¡Hola mundo! Soy Chirp 3", "Bonjour le monde ! Je suis Chirp 3", "Halo dunia! Saya Chirp 3", "Ciao mondo! Sono Chirp 3", "こんにちは世界！私はチャープ3です", "Merhaba dünya! Ben Chirp 3", "Chào thế giới! Tôi là Chirp 3", "হ্যালো ওয়ার্ল্ড! আমি চির্প 3", "હેલો વર્લ્ડ! હું ચિર્પ 3 છું", "ನಮಸ್ಕಾರ ಪ್ರಪಂಚ! ನಾನು ಚಿರ್ಪ್ 3", "ഹലോ വേൾഡ്! ഞാൻ ചിർപ് 3 ആണ്", "नमस्कार जग! मी चिरप 3 आहे", "வணக்கம் உலகம்! நான் சிர்ப் 3", "హలో వరల్డ్! నేను చిర్ప్ 3", "Hallo wereld! Ik ben Chirp 3", "안녕하세요! 저는 Chirp 3입니다", "你好世界！我是 Chirp 3", "Witaj świecie! Jestem Chirp 3", "Привет, мир! Я Чирп 3", "สวัสดีชาวโลก! ฉันคือเชิร์ป 3"]

voice = "Aoede"  # @param ["Aoede", "Puck", "Charon", "Kore", "Fenrir", "Leda", "Orus", "Zephyr"]

language_code = "en-US"  # @param [ "de-DE", "en-AU", "en-GB", "en-IN", "en-US", "fr-FR", "hi-IN", "pt-BR", "ar-XA", "es-ES", "fr-CA", "id-ID", "it-IT", "ja-JP", "tr-TR", "vi-VN", "bn-IN", "gu-IN", "kn-IN", "ml-IN", "mr-IN", "ta-IN", "te-IN", "nl-NL", "ko-KR", "cmn-CN", "pl-PL", "ru-RU", "th-TH"]
# fmt: on

voice_name = f"{language_code}-Chirp3-HD-{voice}"
voice = texttospeech.VoiceSelectionParams(
    name=voice_name,
    language_code=language_code,
)

# Perform the text-to-speech request on the text input with the selected
# voice parameters and audio file type
response = client.synthesize_speech(
    input=texttospeech.SynthesisInput(text=prompt),
    voice=voice,
    # Select the type of audio file you want returned
    audio_config=texttospeech.AudioConfig(
        audio_encoding=texttospeech.AudioEncoding.MP3
    ),
)

display(Audio(response.audio_content))

prompt = """
Google Cloud Text-to-Speech (TTS) is a powerful API that converts text into natural-sounding audio. Here's a breakdown:

**Key Features:**

* **High-Fidelity Speech:** Leverages advanced AI to generate speech that closely resembles human voices.
* **Wide Voice Selection:** Offers a vast library of over 380 voices across 50+ languages and variants, catering to diverse needs.
* **Customization Options:**
    * **Custom Voice:** Create unique voices tailored to your brand or specific requirements using your own audio recordings.
    * **SSML Support:** Utilize Speech Synthesis Markup Language (SSML) to control pronunciation, pacing, and other speech nuances.
* **Integration Flexibility:** Easily integrate with various applications and devices via REST or gRPC APIs.
* **Use Cases:**
    * **Voice Assistants:** Powering conversational AI in smart devices, chatbots, and voice-activated applications.
    * **Accessibility:** Enabling screen readers and text-to-speech features for users with disabilities.
    * **E-learning:** Creating engaging and accessible educational content.
    * **Audiobooks and Podcasts:** Producing high-quality audio for audiobooks, podcasts, and other audio content.
    * **Interactive Experiences:** Enhancing user experiences in games, virtual reality, and other interactive applications.

**In essence, Google Cloud Text-to-Speech empowers developers and businesses to:**

* **Enhance user experiences:** Create more engaging and inclusive interactions through natural-sounding speech.
* **Increase accessibility:** Make information more accessible to a wider audience, including those with visual impairments.
* **Improve efficiency:** Automate tasks like reading aloud documents, generating voiceovers, and creating interactive voice responses.
* **Innovate with new applications:** Explore novel use cases by leveraging the power of AI-powered speech synthesis.

If you're looking to add a voice dimension to your applications or projects, Google Cloud Text-to-Speech is a valuable tool to consider.
"""
final_audio_data = process_streaming_audio(
    prompt, voice, display_individual_chunks=False
)

display(Audio(final_audio_data, rate=24000))