OpenAI TTS

Setup

The OpenAI provider uses the openai npm package as a peer dependency. Install it alongside playwright-recast:

npm install openai

Set your API key as an environment variable:

export OPENAI_API_KEY="sk-..."

Usage

import { Recast } from 'playwright-recast'
import { OpenAIProvider } from 'playwright-recast/providers/openai'

await Recast
  .from('./traces')
  .parse()
  .subtitlesFromSrt('./narration.srt')
  .voiceover(OpenAIProvider({
    voice: 'nova',
    speed: 1.2,
    instructions: 'Professional product demo narration.',
  }))
  .render({ format: 'mp4' })
  .toFile('demo.mp4')

Configuration options

Option	Type	Default	Description
`voice`	`string`	`'nova'`	Voice to use for synthesis
`model`	`string`	`'gpt-4o-mini-tts'`	OpenAI TTS model
`speed`	`number`	`1.0`	Speech speed multiplier
`instructions`	`string`	`undefined`	System prompt for voice style and tone
`apiKey`	`string`	`process.env.OPENAI_API_KEY`	API key (overrides env variable)

Available voices

OpenAI provides six built-in voices:

Voice	Description
`alloy`	Neutral, balanced
`echo`	Warm, conversational
`fable`	Expressive, storytelling
`onyx`	Deep, authoritative
`nova`	Friendly, natural
`shimmer`	Clear, polished

Instructions

The instructions parameter lets you control the voice style and tone. This maps to OpenAI's system-level instructions for TTS:

OpenAIProvider({
  voice: 'nova',
  instructions: 'Calm, professional demo narration. Speak clearly with moderate pacing.',
})

CLI usage

npx playwright-recast -i ./traces --srt narration.srt --provider openai --voice nova
npx playwright-recast -i ./traces --srt narration.srt --provider openai --voice shimmer --tts-speed 1.2

Environment variable

Variable	Required	Description
`OPENAI_API_KEY`	Yes (unless `apiKey` is set)	Your OpenAI API key