Voice and language AI for Bambara

Transcribe, voice and translate Bambara through one API that works with the OpenAI SDKs.

Try the models

Try the models

With a short recording or a few lines of text, in your browser.

Record or import Bambara audio

or drop an audio file in this box

The transcription will appear here.

sunjata-1

Subtitling a full video?Try Djelia Studio

Three models for Bambara

Billed per minute or per character, depending on the model.

sunjata-1

Speech to text

Turns Bambara speech into timestamped text.

Endpoint
/audio/transcriptions
Price
$0.07 / minute
Docs

jifili-1

Text to speech

Reads your Bambara text aloud and streams the audio back.

Endpoint
/audio/speech
Price
$0.06 / 1,000 characters
Docs

banjugu-1

Text translation

Translates between Bambara, French and English, in either direction.

Endpoint
/chat/completions
Price
$0.05 / 1,000 characters
Docs

For people who read, and people who listen

Translation

Djelia Translate

Move a message, a lesson or a notice from French or English into Bambara, and back.

For daily conversations and services with a multilingual public

French

“J’aime apprendre” → “A kadi ne ye ka kalan kɛ”

Translate

English

“Welcome” → “Aw bisimila”

Translate

Bambara

“I ni sogoma” → “Good morning”

Native

Voice

Djelia Voice

An announcement, a lesson or a story, read aloud in Bambara for people who would rather listen.

For radio, schools and creators

Play Moussa
00:00jifili-1 · Moussa00:00

Djelia Console

Try every model without writing code

Test a model in your browser, then create an API key when you are ready to build.

Open console

Track your usage

See how many requests go through, how many fail, and how fast.

Track your usage

Translate side by side

Your text on the left, the translation on the right, your history underneath.

Translate side by side

Djelia API · OpenAI-compatible

Add text to speech to your app.

Three lines change in your code: the key, the base URL and the model name.

Capability
pip install openai
import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["DJELIA_API_KEY"],
    base_url="https://api.djelia.cloud/openai/v1",
)

with client.audio.speech.with_streaming_response.create(
    model="jifili-1",
    input="Aw ni ce, i ka kene wa?",
    voice="moussa",
) as response:
    response.stream_to_file("hello.mp3")
POST/audio/speechhttps://api.djelia.cloud/openai/v1

Create an API key

Keep the key on your server, in DJELIA_API_KEY.

Some OpenAI parameters are ignored. Language pairs go in the djelia object.

Who Djelia is for

Everyday use

Translate a message from French or English into Bambara, or the other way. Have a notice read aloud with jifili-1.

banjugu-1jifili-1

Creators and media

Transcribe a Bambara interview and get its subtitles with Djelia Studio. Then translate them into French or English.

sunjata-1Studio

Education

Translate course material between Bambara, French and English. Add a Bambara audio version for listening in class.

banjugu-1jifili-1

Businesses and services

Transcribe your customers’ Bambara voice messages. Answer in writing or in a Bambara voice.

sunjata-1jifili-1

Where to start

Member of NVIDIA Inception and the Anthropic Startup Program

Common questions

Something else? Write to support@djelia.cloud

API and integration

No. The API is OpenAI-compatible. Keep the official SDK and set the base URL to https://api.djelia.cloud/openai/v1.

Voice

Pricing and access

What will you create in Bambara?

Start in the console. For a team project, write to us.