60db TTS Provider Guide

60db is a hosted speech API with workspace-scoped voices. NeuroLink exposes its HTTP text-to-speech endpoint as sixtydb.

Configuration

Set SIXTYDB_API_KEY to your workspace API key. Supply a workspace voice UUID through tts.voice, or set SIXTYDB_DEFAULT_VOICE. There is no shared default voice. Existing LLM configuration still applies when using NeuroLink.generate().

import {
  NeuroLink,
  TTSProcessor,
  registerDefaultTTSHandlers,
} from "@juspay/neurolink";
import { writeFileSync } from "node:fs";

registerDefaultTTSHandlers();
const voices = await TTSProcessor.getVoices("sixtydb");
if (!voices.length) throw new Error("No voices available in this workspace");

const sdk = new NeuroLink();
const result = await sdk.generate({
  provider: "openai",
  input: { text: "Hello world." },
  tts: {
    enabled: true,
    provider: "sixtydb",
    voice: voices[0].id,
    format: "wav",
  },
});
if (!result.audio) throw new Error("Speech synthesis failed");
writeFileSync("hello.wav", result.audio.buffer);

This synthesizes the input directly. Set tts.useAiResponse: true to synthesize the LLM response instead.

Voice discovery

TTSProcessor.getVoices("sixtydb") loads the workspace's quality and fast catalogs and caches their combined results for five minutes. Pass { languageCode: "hi" } as the second argument to filter using the catalog's language labels. The selected voice determines the synthesis tier; no model ID is sent in the synthesis request.

Audio and options

Option	Supported values
voice	Workspace voice UUID
format	wav (default), pcm16
speed	0.5–2; default 1
Input text	1–5000 characters per synthesis

The handler requests mono LINEAR16 audio at 24 kHz. WAV output includes a RIFF header; pcm16 is raw signed 16-bit little-endian audio. Other formats are rejected before the HTTP request. Native timestamp, pitch and volume controls are not exposed by this handler.

The handler buffers each HTTP synthesis response. stream() uses NeuroLink's existing sentence-based TTS processing; it does not expose 60db's native audio stream. Each request has a 30-second timeout covering headers and response body, and a 64 MiB response limit.

CLI

With the workspace key and default voice configured:

neurolink generate "Hello world." --provider openai --tts \
  --tts-provider sixtydb --tts-format wav --tts-output hello.wav

Use --tts-voice to override the default workspace voice for one request.

Explicit credentials

For applications that manage credentials themselves, register an instance:

import { SixtyDBTTS, TTSProcessor } from "@juspay/neurolink";

TTSProcessor.registerHandler("sixtydb", new SixtyDBTTS(workspaceApiKey));

An optional second constructor argument overrides the API endpoint for an application-managed proxy. Credentials are sent as a Bearer token; HTTP redirects are rejected.

Troubleshooting

• Missing credentials: configure the workspace key before registration.
• Missing voice or invalid UUID: list workspace voices and supply an ID from that catalog.
• Unsupported format: select WAV or PCM16 explicitly, especially in the CLI, whose common format default is MP3.
• Failed generate(): inspect result.ttsMetadata and check result.audio before saving. NeuroLink reports speech failures separately from text output.
• HTTP authentication errors and malformed audio responses are non-retryable.

See the 60db TTS reference and workspace voice reference for the upstream request contract.