Universal-3 Pro Streaming workloads sit on a spectrum between two competing goals: returning transcripts as fast as possible, and returning the most accurate transcripts possible. To make this tradeoff explicit, Universal-3 Pro supports a mode connection parameter you can set when opening a streaming session.
Modes are a Universal-3 Pro Streaming feature. They are not available on Universal-Streaming English or Universal-Streaming Multilingual.
Modes
We offer three modes:
| Mode | Value | When to use |
|---|
| Min latency | min_latency | Lowest possible time-to-text. Best when responsiveness matters more than catching every word. |
| Balanced | balanced | A middle ground between latency and accuracy. Best for voice agents and other interactive applications. |
| Max accuracy | max_accuracy | Highest transcription accuracy. Best for note-taking, scribes, and post-call analysis where a small added delay is acceptable. |
Set the mode
Set the mode connection parameter when you open the WebSocket.
Python
Python SDK
Javascript
JavaScript SDK
CONNECTION_PARAMS = {
"sample_rate": 16000,
"speech_model": "u3-rt-pro",
"mode": "balanced", # min_latency | max_accuracy
}
client.connect(
StreamingParameters(
sample_rate=16000,
speech_model="u3-rt-pro",
mode="balanced", # min_latency | max_accuracy
)
)
const CONNECTION_PARAMS = {
sample_rate: 16000,
speech_model: "u3-rt-pro",
mode: "balanced", // min_latency | max_accuracy
};
const transcriber = client.streaming.transcriber({
sampleRate: 16_000,
speechModel: "u3-rt-pro",
mode: "balanced", // min_latency | max_accuracy
});
