Text-to-Speech API — Python SDK

Official Python wrapper for the Text-to-Speech API on RapidAPI.

Convert text to natural-sounding speech with 400+ voices in 100+ languages, word-level timestamps, adjustable speed and pitch.

Install

pip install text-to-speech-api

Quick Start

from tts_api import TTSAPI

api = TTSAPI("YOUR_RAPIDAPI_KEY")

# Generate speech
audio = api.tts("Hello, world!")
with open("hello.mp3", "wb") as f:
    f.write(audio)

# List French voices
result = api.voices("fr")
for v in result["voices"]:
    print(v["name"], v["friendly_name"])

API

TTSAPI(api_key, base_url=None, timeout=30)

api.health() — Health Check

status = api.health()
# {'status': 'ok', 'voices_count': 322, 'version': '1.0.0'}

api.voices(language=None) — List Voices

# All voices
all_voices = api.voices()

# Spanish voices only
spanish = api.voices("es")

api.tts(text, *, voice, rate, pitch, format) — Text to Speech

Returns audio as bytes. Free tier.

audio = api.tts(
    "Breaking news from around the world.",
    voice="en-US-GuyNeural",
    rate="+20%",
    format="mp3",
)
with open("news.mp3", "wb") as f:
    f.write(audio)

api.tts_with_timestamps(text, *, voice, rate, pitch, format) — TTS + Timestamps

Returns audio as base64 + word-level timing. Basic tier.

result = api.tts_with_timestamps("Five amazing facts about space.")
for word in result["timestamps"]:
    print(f"{word['text']} at {word['offset_ms']}ms")

# Decode audio
audio = api.decode_audio(result["audio_base64"])
with open("space.mp3", "wb") as f:
    f.write(audio)

Pricing

Requirements

• Python 3.8+
• Zero dependencies (uses only stdlib)

License

MIT