SpeakEasy Node.js SDK

Official Node.js/TypeScript client for the SpeakEasy API — affordable speech-to-text and text-to-speech for developers.

Installation

npm install tryspeakeasy

Quick Start

import { SpeakEasy } from 'tryspeakeasy'

const se = new SpeakEasy('sk-se-your_api_key')

// Transcribe audio
const { text } = await se.transcribe({
  url: 'https://example.com/audio.mp3',
})
console.log(text)

// Generate speech
const audio = await se.speech({
  input: 'Hello world!',
  voice: 'nova',
})

Get your API key from the SpeakEasy dashboard.

Usage

Speech-to-Text

Transcribe audio files or URLs. Supports mp3, wav, flac, aac, opus, ogg, m4a, mp4, and webm (max 100 MB).

// From URL
const result = await se.transcribe({
  url: 'https://example.com/podcast.mp3',
})
console.log(result.text)

// From file (Node.js)
import { readFileSync } from 'fs'
const file = new Blob([readFileSync('recording.mp3')])
const result = await se.transcribe({ file })

// With options
const result = await se.transcribe({
  url: 'https://example.com/audio.mp3',
  language: 'en',
  speakerLabels: true,
  wordTimestamps: true,
  responseFormat: 'verbose_json',
})
console.log(result.segments) // timestamped segments

Response Formats

// JSON (default) — { text: string }
const json = await se.transcribe({ url, responseFormat: 'json' })

// Verbose JSON — includes segments with timestamps
const verbose = await se.transcribe({ url, responseFormat: 'verbose_json' })

// Plain text
const text = await se.transcribe({ url, responseFormat: 'text' })

// Subtitles
const srt = await se.transcribe({ url, responseFormat: 'srt' })
const vtt = await se.transcribe({ url, responseFormat: 'vtt' })

Async with Webhook

For long audio files, provide a callbackUrl to get results asynchronously:

await se.transcribe({
  url: 'https://example.com/long-podcast.mp3',
  callbackUrl: 'https://your-server.com/webhook',
})
// Your server receives the result via POST when done

Text-to-Speech

Convert text to audio with 40+ voices across 10 languages.

// Generate speech (returns ArrayBuffer)
const audio = await se.speech({
  input: 'Welcome to SpeakEasy!',
  voice: 'nova',
})

// Save to file (Node.js)
import { writeFileSync } from 'fs'
writeFileSync('output.mp3', Buffer.from(audio))

// With options
const audio = await se.speech({
  input: 'Bonjour le monde!',
  voice: 'amelie',
  language: 'fr',
  responseFormat: 'wav',
  speed: 1.2,
})

// With word timestamps (returns JSON instead of audio)
const result = await se.speech({
  input: 'Hello world',
  wordTimestamps: true,
})
console.log(result.word_timestamps)
// [{ word: 'Hello', start: 0.0, end: 0.4 }, { word: 'world', start: 0.5, end: 0.9 }]

Available Voices

| Language | Voices | |----------|--------| | English (US) | heart, bella, michael, alloy, aoede, kore, jessica, nicole, nova, river, sarah, sky, echo, eric, fenrir, liam, onyx, puck, adam, santa | | English (UK) | alice, emma, isabella, lily, daniel, fable, george, lewis | | Japanese | haruto, yuki | | Chinese | xiaobei, yunjian | | Spanish | carlos, maria | | French | pierre, amelie | | Hindi | arjun, priya | | Italian | luca, giulia | | Portuguese (BR) | pedro, ana |

Error Handling

import { SpeakEasy, SpeakEasyError } from 'tryspeakeasy'

try {
  await se.transcribe({ url: 'https://example.com/audio.mp3' })
} catch (err) {
  if (err instanceof SpeakEasyError) {
    console.log(err.status)  // 402, 429, etc.
    console.log(err.message) // 'No active subscription'
  }
}

| Status | Meaning | |--------|---------| | 401 | Missing or invalid API key | | 403 | No active subscription | | 429 | Too many concurrent requests (max 10) |

Configuration

const se = new SpeakEasy('sk-se-your_api_key', {
  // Override the base URL
  baseUrl: 'https://staging.tryspeakeasy.io',
  // Provide a custom fetch implementation
  fetch: myCustomFetch,
})

API Reference

| Method | Description | |--------|-------------| | transcribe(options) | Speech-to-text — transcribe audio files or URLs | | speech(options) | Text-to-speech — generate audio from text |

Requirements

• Node.js 18+ (uses native fetch)
• A SpeakEasy account with an active subscription

Links

• SpeakEasy — Affordable Speech-to-Text & Text-to-Speech API
• Dashboard — API key management and usage monitoring
• Documentation — Full API reference
• Pricing — Usage-based pricing starting at $9/month

License

MIT

Built by SIÁN Agency