Pre-Recorded Audio (STT)

Transcribe uploaded audio files with advanced features like speaker diarization and multi-language support. Our API follows OpenAI and Deepgram compatible patterns.

SDK integration

Use the official Deepgram SDK with our API endpoint.

import { createClient } from '@deepgram/sdk';

const client = createClient('YOUR_API_KEY', {
  global: {
    fetch: { options: { url: 'https://api.greenpt.ai/v1' } },
  },
});

from deepgram import DeepgramClient
from deepgram.environment import DeepgramClientEnvironment

# Create environment for GreenPT API
greenpt_env = DeepgramClientEnvironment(
    base="https://api.greenpt.ai",
    production="wss://api.greenpt.ai",
    agent="wss://api.greenpt.ai",
)

deepgram = DeepgramClient("YOUR_API_KEY", environment=greenpt_env)

Endpoint

POST https://api.greenpt.ai/v1/listen

Request body

Required and optional parameters.

Example request: local file

curl \
  --request POST \
  --header 'Authorization: Token YOUR_API_KEY' \
  --header 'Content-Type: audio/wav' \
  --data-binary @youraudio.wav \
  --url 'https://api.greenpt.ai/v1/listen?model=green-s&language=en&diarize_model=v2&punctuate=true'

Example request: URL / bucket

curl \
  --request POST \
  --header 'Authorization: Token YOUR_API_KEY' \
  --header 'Content-Type: application/json' \
  --data '{"url":"https://static.deepgram.com/examples/Bueller-Life-moves-pretty-fast.wav"}' \
  --url 'https://api.greenpt.ai/v1/listen?model=green-s&language=en&diarize_model=v2&punctuate=true'

Example response

{
  "metadata": {
    "request_id": "a847f427-4ad5-4d67-9b95-db801e58251c",
    "duration": 25.933313,
    "channels": 1,
    "created": "2024-05-12T18:57:13.426Z"
  },
  "results": {
    "channels": [
      {
        "alternatives": [
          {
            "transcript": "Hi, thanks for calling support. Yeah, my account is locked.",
            "confidence": 0.98,
            "words": [
              { "word": "hi", "start": 0.5, "end": 0.8, "confidence": 0.99, "speaker": 0 },
              { "word": "thanks", "start": 0.8, "end": 1.1, "confidence": 0.98, "speaker": 0 },
              { "word": "for", "start": 1.1, "end": 1.25, "confidence": 0.99, "speaker": 0 },
              { "word": "calling", "start": 1.25, "end": 1.6, "confidence": 0.98, "speaker": 0 },
              { "word": "support", "start": 1.6, "end": 2.1, "confidence": 0.97, "speaker": 0 },
              { "word": "yeah", "start": 2.4, "end": 2.7, "confidence": 0.98, "speaker": 1 },
              { "word": "my", "start": 2.7, "end": 2.85, "confidence": 0.99, "speaker": 1 },
              { "word": "account", "start": 2.85, "end": 3.2, "confidence": 0.98, "speaker": 1 },
              { "word": "is", "start": 3.2, "end": 3.35, "confidence": 0.99, "speaker": 1 },
              { "word": "locked", "start": 3.35, "end": 3.8, "confidence": 0.97, "speaker": 1 }
            ]
          }
        ]
      }
    ]
  }
}

Complete SDK example

Full working example with the Deepgram SDK.

import { createClient } from '@deepgram/sdk';
import fs from 'fs';

const client = createClient('YOUR_API_KEY', {
  global: {
    fetch: { options: { url: 'https://api.greenpt.ai/v1' } },
  },
});

async function transcribeLocalFile() {
  const { result, error } = await client.listen.prerecorded.transcribeFile(
    fs.readFileSync('path/to/audio.wav'),
    {
      model: 'green-s',
      language: 'en',
      punctuate: true,
      diarize_model: 'v2',
    },
  );

  if (error) throw error;
  console.log(result);
}

transcribeLocalFile();

from deepgram import DeepgramClient
from deepgram.environment import DeepgramClientEnvironment

# Create environment for GreenPT API
greenpt_env = DeepgramClientEnvironment(
    base="https://api.greenpt.ai",
    production="wss://api.greenpt.ai",
    agent="wss://api.greenpt.ai",
)

deepgram = DeepgramClient("YOUR_API_KEY", environment=greenpt_env)

# For URL-based transcription
response = deepgram.listen.v1.media.transcribe_url(
    url="https://static.deepgram.com/examples/Bueller-Life-moves-pretty-fast.wav",
    model="green-s",
    language="en",
    punctuate=True,
    diarize_model="v2",
    smart_format=True,
)

print(response)

Rate limits

Requests are limited per account (shared across all your API keys): 600 requests per 5-minute window, and shared across every GreenPT API endpoint rather than counted separately per endpoint. Exceeding it returns 429 Too many requests, please try again later. with RateLimit-Limit, RateLimit-Remaining, RateLimit-Reset, and Retry-After response headers — read RateLimit-Remaining to track your usage against the current window.

Available models

Choose the model that fits your language and use case.

green-s: GreenS

Reliable speech-to-text for single-language audio. Great for recordings, podcasts, and archived content.

Supported languages: English, German, Spanish, French, Italian, Dutch, Portuguese, Romanian, Bulgarian, Catalan, Danish, Finnish, Swedish.

green-s-pro: GreenS Pro

Advanced model with automatic language detection. Ideal for international content and mixed-language recordings.

Supported languages: English, German, Dutch, Swedish, Turkish.

Multilingual: use multi for automatic language detection across languages in the same file.

Multilingual mode

Transcribe conversations where speakers switch between languages.

With green-s-pro, set language=multi to transcribe audio where multiple languages are spoken. The model automatically detects and transcribes each language as speakers switch.

Languages supported in multilingual mode: English, Spanish, French, German, Hindi, Russian, Portuguese, Japanese, Italian, Dutch.

Price difference

Multilingual processing costs more than single-language. For pre-recorded: €0.28/hour vs €0.23/hour for monolingual.

Output differences

With language=multi, the response adds a languages array and a language field per word:

"alternatives": [{
  "transcript": "No recuerdo mi bank password.",
  "languages": ["es", "en"],
  "words": [
    { "word": "no", "language": "es" },
    { "word": "recuerdo", "language": "es" },
    { "word": "bank", "language": "en" }
  ]
}]

Speaker diarization

Diarization labels which speaker said each word. Use diarize_model to enable it and pick the model version in one parameter; you don't need to also set diarize:

Diarization V2 improves speaker-attribution accuracy and reduces labeling errors. It works across all pre-recorded STT models (green-s, green-s-pro) and languages at no extra cost.

The legacy diarize=true flag still works but always uses v1. Do not set both diarize and diarize_model; requests that include both are rejected. Diarization V2 is not available for live streaming.

Available features

Add-on capabilities for pre-recorded transcription.

Pricing

Pre-recorded transcription rates per hour of audio.

All prices in EUR, excl. taxes. green-s and green-s-pro monolingual share the same pre-recorded rate. Choose green-s-pro for multilingual mode or additional language options.

Additional features

Launch promo: these add-ons are free right now. The prices shown are the standard per-hour rates that apply once the promo ends; you are not charged for them today.

Speaker diarization is included free for pre-recorded audio. Multichannel audio is billed per channel, so 2-channel audio is charged at double the per-hour rate.