Skip to main content

​
Getting started with Timekettle Developer

Tmk Translation SDK packages Timekettle’s real-time speech translation capabilities for mobile applications. It provides an end-to-end translation pipeline that combines automatic speech recognition, machine translation, large language model translation, and text-to-speech output.
Exact language coverage, offline model availability, credentials, and package access depend on your Timekettle agreement and SDK distribution channel.

​
What the SDK provides

Full speech translation pipeline

Capture speech input, recognize spoken text, translate it, and synthesize translated speech output in one SDK workflow.

Android and iOS support

Integrate Timekettle translation capabilities into native Android and iOS applications.

Online translation mode

Use cloud-backed translation for broad language and accent coverage when network connectivity is available.

Offline translation mode

Support selected language pairs for translation scenarios where network access is limited or unavailable.

​
Core capabilities

The SDK is designed for applications that need real-time voice translation rather than isolated text translation. Its core capabilities include:
  • Automatic speech recognition: Convert spoken audio into recognized text.
  • Machine translation and LLM translation: Translate recognized text into the target language.
  • Text-to-speech: Generate translated speech output for playback.
  • PCM audio workflow: Process audio input and output as 16 kHz, 16-bit PCM.
  • Online language coverage: Support broad online language and accent coverage for cloud-backed translation.
  • Offline language coverage: Support selected offline translation directions for no-network or low-connectivity scenarios.

​
Translation modes

Online mode is intended for production experiences that need the broadest language and accent coverage. It can use Timekettle cloud services for speech recognition, translation, text-to-speech, room connection, messaging, and diagnostics.

​
Before you integrate

Decide whether your first integration targets Android, iOS, or both. Platform-specific setup, permissions, packaging, and release checks should be handled in the corresponding integration guide.
If your app records microphone audio, request platform microphone permission in your own app and explain the recording purpose in your user-facing privacy notice.
Review Privacy Policy, Terms of Service, and any SDK-specific data handling guidance before enabling diagnostics or processing end-user audio.
Confirm that your required source and target languages are supported by the selected translation mode before committing to a release plan.

​
Next steps

Review privacy requirements

Understand how developer data, end-user audio, translated text, diagnostics, and SDK storage may be processed.

Review service terms

Review API, SDK, acceptable use, service availability, and developer responsibility terms.