Speech Synthesis

Convert text to natural-sounding speech using TTS models.

Endpoint

POST /v1/audio/speech

Authentication

Authorization: Bearer gw_prod_...

Request

Headers

Header	Required	Description
Authorization	Yes	Bearer token with API key
Content-Type	Yes	application/json

Body Parameters

Parameter	Type	Required	Description
model	string	Yes	Model ID (see supported models)
input	string	Yes	Text to synthesize (max 4096 chars)
voice	string	Yes	Voice ID or standard voice name
response_format	string	No	mp3, opus, aac, flac, wav, pcm
speed	number	No	Speed multiplier (0.25-4.0, default 1.0)

Examples

Basic Speech

curl -X POST https://api.gateflow.ai/v1/audio/speech \
  -H "Authorization: Bearer gw_prod_..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "tts-1",
    "input": "Hello, welcome to GateFlow!",
    "voice": "alloy"
  }' --output speech.mp3

Python

import openai

client = openai.OpenAI(
    base_url="https://api.gateflow.ai/v1",
    api_key="gw_prod_..."
)

response = client.audio.speech.create(
    model="tts-1",
    input="Hello, welcome to GateFlow!",
    voice="alloy"
)

response.stream_to_file("speech.mp3")

High Quality

response = client.audio.speech.create(
    model="tts-1-hd",
    input="Hello, welcome to GateFlow!",
    voice="nova"
)

ElevenLabs Premium

response = client.audio.speech.create(
    model="eleven_multilingual_v2",
    input="Hello, welcome to GateFlow!",
    voice="rachel"
)

Using Standard Voices

# Standard voice maps to provider-specific voice
response = client.audio.speech.create(
    model="eleven_turbo_v2_5",
    input="How can I help you today?",
    voice="friendly"  # Maps to "rachel" on ElevenLabs
)

Streaming Speech

import httpx

async with httpx.AsyncClient() as http_client:
    async with http_client.stream(
        "POST",
        "https://api.gateflow.ai/v1/audio/speech",
        headers={
            "Authorization": "Bearer gw_prod_...",
            "Content-Type": "application/json"
        },
        json={
            "model": "eleven_turbo_v2_5",
            "input": "This is streaming speech synthesis.",
            "voice": "friendly",
            "stream": True
        }
    ) as response:
        async for chunk in response.aiter_bytes():
            play_audio_chunk(chunk)

Custom Voice Settings

response = client.audio.speech.create(
    model="eleven_multilingual_v2",
    input="Hello!",
    voice="friendly",
    extra_body={
        "gateflow": {
            "voice_settings": {
                "stability": 0.5,
                "similarity_boost": 0.75,
                "style": 0.5,
                "speed": 1.0
            }
        }
    }
)

Response

The response is an audio file in the requested format.

Headers

Header	Description
Content-Type	audio/mpeg, audio/opus, etc.
Content-Length	Size in bytes
X-GateFlow-Provider	Provider used
X-GateFlow-Model	Model used
X-GateFlow-Latency-Ms	Processing time

Supported Models

OpenAI

Model	Quality	Latency	Cost
tts-1	Good	Low	$15/1M chars
tts-1-hd	High	Medium	$30/1M chars

ElevenLabs

Model	Quality	Latency	Cost
eleven_multilingual_v2	Excellent	Medium	$24/1M chars
eleven_turbo_v2_5	Good	Low	$12/1M chars
eleven_flash_v2_5	Good	Very Low	$8/1M chars

Standard Voices

Voice ID	Persona	OpenAI	ElevenLabs
professional	Business	onyx	josh
friendly	Conversational	alloy	rachel
calm	Soothing	nova	bella
energetic	Upbeat	shimmer	antoni
serious	Formal	fable	adam
casual	Relaxed	echo	domi

GateFlow Extensions

Provider Selection

response = client.audio.speech.create(
    model="auto",
    input="Hello!",
    voice="friendly",
    extra_body={
        "gateflow": {
            "prefer": "quality"  # or "speed" or "cost"
        }
    }
)

Fallbacks

response = client.audio.speech.create(
    model="eleven_multilingual_v2",
    input="Hello!",
    voice="friendly",
    extra_body={
        "gateflow": {
            "fallbacks": ["tts-1-hd", "tts-1"]
        }
    }
)

Audio Formats

Format	MIME Type	Use Case
mp3	audio/mpeg	General use, good compression
opus	audio/opus	Streaming, low latency
aac	audio/aac	Apple devices
flac	audio/flac	Lossless, archival
wav	audio/wav	Uncompressed, editing
pcm	audio/pcm	Raw audio, processing

Errors

Code	Description
400	Invalid parameters or text too long
401	Invalid API key
429	Rate limit exceeded
500	Provider error

See Also

• Transcriptions - Speech to text
• Voices - List available voices
• Voice Mapping - Voice configuration