Latency optimization
This guide shows you how to reduce text-to-speech latency in your application.
This guide covers the core principles for improving text-to-speech latency. For a conceptual explanation of what latency is and what contributes to it, see Understanding latency.
While there are many individual techniques, we’ll group them into four principles.
Four principles
Section titled “Four principles”Use Flash models
Section titled “Use Flash models”Flash models deliver ~75ms inference speeds, making them ideal for real-time applications. The trade-off is a slight reduction in audio quality compared to Multilingual v2.
Leverage streaming
Section titled “Leverage streaming”There are three types of text-to-speech endpoints available in our API Reference:
- Regular endpoint: Returns a complete audio file in a single response.
- Streaming endpoint: Returns audio chunks progressively using Server-sent events.
- Websockets endpoint: Enables bidirectional streaming for real-time audio generation.
Streaming
Section titled “Streaming”Streaming endpoints progressively return audio as it is being generated in real-time, reducing the time-to-first-byte. This endpoint is recommended for cases where the input text is available up-front.
Websockets
Section titled “Websockets”The text-to-speech websocket endpoint supports bidirectional streaming making it perfect for applications with real-time text input (e.g. LLM outputs).
If auto_mode is disabled, the model will wait for enough text to match the chunk schedule before starting to generate audio.
For instance, if you set a chunk schedule of 125 characters but only 50 arrive, the model stalls until additional characters come in—potentially increasing latency.
For implementation details, see the text-to-speech websocket guide.
Choose appropriate voices
Section titled “Choose appropriate voices”We have observed that in some cases, voice selection can impact latency. Here’s the order from fastest to slowest:
- Default voices (formerly premade), Synthetic voices, and Instant Voice Clones (IVC)
- Professional Voice Clones (PVC)
Higher audio quality output formats can increase latency. Be sure to balance your latency requirements with audio fidelity needs.
Consider geographic proximity
Section titled “Consider geographic proximity”We serve our models from multiple regions to optimize latency based on your geographic location.
For example, using Flash models with Websockets, you can expect the following TTFB latencies depending on your location:
| Region | TTFB |
|---|---|
| North America | 100-150ms |
| Europe | 100-150ms |
| South East Asia | 100-150ms |
| South Asia | 150-200ms |
| North East Asia | 150-200ms |
You can check which backend region is serving your request by inspecting the x-region header in the API response. Currently used regions include: USA, Netherlands and Singapore.
To opt-out of the global routing and always use USA servers, use the api.us.elevenlabs.io base URL for your API requests:
curl -X POST -v "https://api.us.elevenlabs.io/v1/text-to-speech/{voice_id}" \
-H "Accept: audio/mpeg" \
-H "Content-Type: application/json" \
-H "xi-api-key: YOUR_API_KEY" \
-d '{
"text": "Hello World!",
"model_id": "eleven_flash_v2_5"
}'The global servers were previously opt-in using the api-global-preview.elevenlabs.io base URL.
This is no longer necessary as it is now the default behavior. Please update your applications to
simply use api.elevenlabs.io instead.
import os
from elevenlabs.client import ElevenLabs
elevenlabs = ElevenLabs(
api_key=os.getenv("ELEVENLABS_API_KEY"),
base_url="https://api.us.elevenlabs.io"
)