AI audio translator with speech-to-text, LLM translation, and text-to-speech

Upload any audio (podcast, meeting, lecture), STT transcribes in source language, AI Inference translates, TTS generates audio in target language. Returns translated audio + aligned transcript.

STT Transcribe:POST /v2/ai/transcribe

--refAI Inference:POST /v2/ai/chat/completions

--refTTS Generate:POST /v2/ai/generate

--ref

  API Request
        │
        ▼
  ┌──────────────────┐
  │ Answer + Greet    │ ── TTS welcome message
  └────────┬─────────┘
           │
           ▼
  ┌──────────────────┐
  │ Gather Speech     │ ── STT transcription
  └────────┬─────────┘
           │
           ▼
  ┌──────────────────┐
  │ AI Inference      │
  │ • Translation      │
  └────────┬─────────┘
           │ ◄──── conversation loop
           │
           ▼
     JSON response

• Sends conversation to Telnyx AI Inference for processing
• Converts response to speech via Telnyx TTS

Telnyx is an AI Communications Infrastructure platform - voice, messaging, SIP, AI, and IoT on one private, global network.

Co-located inference- LLM runs on the same network as voice traffic. Sub-200ms round trips.

Copy .env.example

to .env

and fill in:

AI_MODEL

string

moonshotai/Kimi-K2.6

DocsTTS_MODEL

string

telnyx/tts

DocsSTT_MODEL

string

telnyx/asr

Docs

git clone https://github.com/team-telnyx/telnyx-code-examples.git
cd telnyx-code-examples/ai-content-translator-python
cp .env.example .env
pip install -r requirements.txt
python app.py
ngrok http 5000

Set webhook URL in Telnyx Portal:

• Call Control Application -> https://<id>.ngrok.io/webhooks/voice

Upload as multipart form:

curl -X POST http://localhost:5000/translate \
  -F audio=@lecture.mp3 \
  -F source=en \
  -F target=ja

Response:

{"job_id": "tr-a1b2c3d4", "status": "complete", "source": "en (English)", "target": "ja (Japanese)", "original_length": 1847, "translated_length": 923}
curl http://localhost:5000/health
{"status": "ok"}

Connection refused on port 5000: App isn't running. Runpython app.py

and check no other process uses port 5000.401 Unauthorized: YourTELNYX_API_KEY

is invalid. Generate a new one atportal.telnyx.com/api-keys.AI response slow/empty: Verify model name. See available models atdevelopers.telnyx.com.

run-llm-inference-python- Standalone inferencebuild-voice-ai-agent-python- Voice AI agent