Signal Tank field guide

Listen with
a method.

A practical route from your first generated line to a useful product decision.

Quick start

The Lab guides you through four numbered steps, so your first useful comparison starts with a clear baseline.

1
Script
Use Narration unchanged or write your own line. Expression tags can guide warmth, emphasis, and pauses.
2
Voice
Start with Balanced, choose another profile, or record an approved clone. The recorder shows whether your microphone is receiving sound.
3
Delivery
Choose speed, volume, latency, format, and variation. Defaults are already suitable for a first take.
4
Review and generate
Check the summary, generate the signal, then compare audio, first-byte time, completion time, and size under Session takes.

Choose a profile or clone

Voice	Character	Useful for
Balanced	Natural and flexible	A clean baseline
Female · American	Bright and friendly	Conversation and narration
Male · American	Clear and energetic	Promotional delivery
Female · British	Crisp and professional	Education and explainers
Male · British	Deep and measured	Authoritative narration
Male · Indian English	Calm and clear	Professional speech
Voice clone	Your approved speaker	A reference clip and exact transcript

For voice cloning

Record in the browser or upload clean audio with one speaker. Aim for 15–20 seconds; the minimum is 10 seconds, the recorder stops at 30, and every clip must be 4 MB or smaller.

Microphone access is requested only after you press Start recording or select the sonar orb. While recording, select the orb again or use Stop recording. The clip stays in browser memory until you press Generate. If recording is unavailable or blocked, switch to Upload audio.

A small test plan that tells you something

Naturalness

Use a paragraph with short and long sentences, names, numbers, and at least one emotional direction.

Agent response

Use the conversational preset at normal and low latency. Listen for pace as well as total response time.

Multilingual

Test content written by a fluent speaker. Check pronunciation, rhythm, and whether expression survives translation.

Delivery range

Hold the voice constant while changing speed, temperature, and diversity in small increments.

Run it locally

Create .env in the repository root—the same folder as package.json.

SIGNAL_TANK_API_KEY=your_voice_api_key
PORT=3000

Then install and start the project:

npm install
npm start

Open 127.0.0.1:3000. The browser never receives the API key.

Call the local interface

The lab posts multipart form data to the server. A minimal browser request looks like this:

const form = new FormData();
form.set("text", "Hello from Signal Tank.");
form.set("voiceMode", "preset");
form.set("voicePreset", "balanced");
form.set("format", "mp3");
form.set("latency", "balanced");

const response = await fetch("/api/tts", {
  method: "POST",
  body: form,
});

const audio = new Audio(URL.createObjectURL(await response.blob()));
audio.play();

Available preset values are balanced, female-american, male-american, female-british, male-british, and male-indian. Signal Tank keeps their internal mapping and credentials on the server.

Privacy and consent

Signal Tank does not persist scripts, generated audio, uploaded references, or browser recordings. Recorded clips stay in memory until generation and generated takes disappear when the page refreshes.

An upstream processing partner may retain requests. Do not submit confidential text, and clone a voice only when the speaker has explicitly approved that use.