Skip to main content

Google Cloud TTS

Google Cloud Text-to-Speech provides high-quality voice synthesis with support for multiple languages, voices, and neural models.

Authentication​

Google Cloud TTS requires a service account JSON file. You can provide it in two ways:

from tts_wrapper import GoogleTTS, GoogleClient

# Method 1: Path to service account JSON file
client = GoogleClient(credentials='path/to/creds.json')

# Method 2: Service account credentials as dictionary
import json
with open('path/to/creds.json', 'r') as file:
    credentials_dict = json.load(file)
client = GoogleClient(credentials=credentials_dict)

tts = GoogleTTS(client)
tip

Use environment variables for secure credential management:

import os
client = GoogleClient(credentials=os.getenv('GOOGLE_SA_PATH'))

Features​

SSML Support​

Google Cloud TTS provides comprehensive SSML support:

ssml_text = """
<speak>
    Hello <break time="300ms"/> World!
    <prosody rate="slow" pitch="+2st">
        This is a test of SSML features.
    </prosody>
</speak>
"""
tts.speak(ssml_text)

See Google Cloud TTS SSML Reference for all supported tags.

Streaming​

Supports real-time audio streaming:

# Stream synthesis for real-time playback
tts.speak_streamed("This text will be synthesized and played in real-time")

Word Timing​

Get precise timing information for each word:

def on_word(word: str):
    print(f"Speaking: {word}")

tts.connect("started-word", on_word)
tts.speak("This text will trigger word timing callbacks")

Voice Selection​

List and select from available voices:

# Get list of available voices
voices = tts.get_voices()
for voice in voices:
    print(f"Name: {voice['name']}")
    print(f"Languages: {voice['language_codes']}")
    print(f"Gender: {voice['gender']}")
    print("---")

# Set a specific voice and language
tts.set_voice("en-US-Standard-C", "en-US")

Neural Voices​

Google Cloud TTS offers high-quality neural voices:

# Select a neural voice
tts.set_voice("en-US-Neural2-C", "en-US")

Best Practices​

  1. Cost Management

    • Use caching for frequently used phrases
    • Monitor usage through Google Cloud Console
    • Consider using standard voices for development

  2. Performance

    • Reuse client instances
    • Use appropriate audio profiles
    • Choose the closest region for lower latency

  3. Error Handling

    try:
        tts.speak("Hello, world!")
    except Exception as e:
        if "InvalidCredentials" in str(e):
            print("Check your Google Cloud credentials")
        elif "QuotaExceeded" in str(e):
            print("Usage quota exceeded")
        else:
            print(f"Error: {e}")

Limitations​

  • Maximum text length of 5000 characters per request
  • API rate limits apply (check Google Cloud quotas)
  • Neural voices may have higher latency
  • Some SSML features are voice-specific
  • Pricing varies by voice type (standard vs. neural)

Additional Resources​

Next Steps​