Echo voices and Baobab national models — Africa-first language intelligence.

Explore Baobab
Skip to content
Open consoleLog in

Speech-to-text

Lugemi Speech

Transcribe accents, dialects, and code-switching built for African and global languages.

Open SpeechCoverage

Lugemi Speech turns spoken audio into text that respects regional phrasing. Pair STT with Translate and Agents for end-to-end speaking-agent pipelines.

Speech → agent reply
Habari, naomba msaada kwa lugha yangu.
Karibu. Ninaweza kusikiliza na kujibu kwa Kiswahili au Kiingereza.

Play each turn, or play the full agent conversation.

Use cases, demos & developer kit

How teams ship with Lugemi Speech: interactive demos above, step guides below, and the same /v1 surface for APIs and SDKs.

Use case guide

Speech guide

Upload audio, transcribe, then hand off to Translate or Agents.

  1. Use the Speech console or Speech Recognition.
  2. Pair transcripts with Translate for captions and dubbing scripts.
  3. Ship speaking agents that listen and reply aloud.
Speech consoleAPI docsSDKs

API & SDK

Speech API & SDK

Recognize audio, then chain translate or voice simulate on the same /v1 surface.

  1. POST audio to Speech Recognition with your workspace key.
  2. Pipe text into Translate or Agents for bilingual replies.
  3. Use @lugemi/sdk helpers or raw OpenAPI clients.
OpenAPI docsPlaygroundSDKs & CLI
Speech consoleAPI docsSDKsOpenAPI docsPlaygroundSDKs & CLI

Transcription

Accents, dialects, and switches are first-class — not edge cases.

Agent pipelines

STT feeds FAQ and agent runtimes so replies can go back out as voice.

Africa's AI communication platform

Start creating with Lugemi Voice, Speech, and Translate — or talk to us about your workspace.

Start buildingTalk to us
Lugemi Speech · Lugemi