Name: SeamlessM4T API: Speech and Text Translation by Meta
Brand: magicapi
SKU: api-magicapi-multilingual-multimodal-translation-cjwbw-seamless-communication
Availability: InStock
Author: API Market

Shashank Agarwal

# 🌐 Speech and Text Translation API (Meta SeamlessM4T)

## Overview
SeamlessM4T is a translation model from Meta. Its developer describes it as massively multilingual and multimodal machine translation. This listing runs the version published by cjwbw.

Choose one of five tasks with `task_name`: speech to speech translation (S2ST, default), speech to text translation (S2TT), text to speech translation (T2ST), text to text translation (T2TT) or automatic speech recognition (ASR). Send `input_audio` for speech-input tasks or `input_text` (with `input_text_language`) for text-input tasks, and pick the target language. Each request returns an object with `text_output` and, for speech-output tasks, an `audio_output` URL.

## Features
- **Five tasks**: S2ST, S2TT, T2ST, T2TT and ASR, selected with `task_name`.
- **Many text languages**: 95 languages for text output with `target_language_text_only` (default Norwegian Nynorsk).
- **Speech output in 36 languages**: `target_language_with_speech` (default French) for S2ST and T2ST.
- **Text and audio together**: speech-output tasks return the translated text as well as the audio.
- **Input length cap**: `max_input_audio_length` (default 60) limits how much input audio is used.

## Use Cases
1. **Dubbing short clips**: turn a spoken clip into speech in another language (S2ST).
2. **Subtitles in another language**: translate speech straight to text (S2TT).
3. **Spoken announcements**: turn written text into speech in another language (T2ST).
4. **Text translation**: translate text between many languages (T2TT).
5. **Transcription**: convert speech to text (ASR).

## Visual Examples
### Example Input
A speech to speech translation (S2ST) of a short English speech clip into French. `target_language_text_only` and `input_text_language` are not used by S2ST and were left at their defaults:

[Listen to the input audio](https://blog.api.market/wp-content/uploads/2026/10/seamless_communication-input-api-market.mp3)

```json
{
  "task_name": "S2ST (Speech to Speech translation)",
  "input_audio": "https://blog.api.market/wp-content/uploads/2026/10/seamless_communication-input-api-market.mp3",
  "input_text_language": "None",
  "max_input_audio_length": 60,
  "target_language_text_only": "Norwegian Nynorsk",
  "target_language_with_speech": "French"
}
```

### Example Output
The request returned French speech and its text:

- `text_output`: "Le modèle M4T sans faille de MetaAI démocratise la communication parlée à travers les barrières linguistiques."
- `audio_output`: [Listen to the example output](https://blog.api.market/wp-content/uploads/2026/10/seamless_communication-output-21w3z6vb-api-market.wav)

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `task_name` | string | No | `S2ST (Speech to Speech translation)` | `S2ST (Speech to Speech translation)`, `S2TT (Speech to Text translation)`, `T2ST (Text to Speech translation)`, `T2TT (Text to Text translation)`, `ASR (Automatic Speech Recognition)` | Task to run |
| `input_audio` | string (URL) | No | none |  | Input audio URL for tasks with speech input: S2ST, S2TT and ASR. |
| `input_text` | string | No | none |  | Input text for tasks with text input: T2ST and T2TT. |
| `input_text_language` | string | No | `None` | `None`, `Afrikaans`, `Amharic`, `Armenian`, `Assamese`, `Basque`, `Belarusian`, `Bengali`, `Bosnian`, `Bulgarian`, `Burmese`, `Cantonese`, `Catalan`, `Cebuano`, `Central Kurdish`, `Croatian`, `Czech`, `Danish`, `Dutch`, `Egyptian Arabic`, `English`, `Estonian`, `Finnish`, `French`, `Galician`, `Ganda`, `Georgian`, `German`, `Greek`, `Gujarati`, `Halh Mongolian`, `Hebrew`, `Hindi`, `Hungarian`, `Icelandic`, `Igbo`, `Indonesian`, `Irish`, `Italian`, `Japanese`, `Javanese`, `Kannada`, `Kazakh`, `Khmer`, `Korean`, `Kyrgyz`, `Lao`, `Lithuanian`, `Luo`, `Macedonian`, `Maithili`, `Malayalam`, `Maltese`, `Mandarin Chinese`, `Marathi`, `Meitei`, `Modern Standard Arabic`, `Moroccan Arabic`, `Nepali`, `North Azerbaijani`, `Northern Uzbek`, `Norwegian Bokmål`, `Norwegian Nynorsk`, `Nyanja`, `Odia`, `Polish`, `Portuguese`, `Punjabi`, `Romanian`, `Russian`, `Serbian`, `Shona`, `Sindhi`, `Slovak`, `Slovenian`, `Somali`, `Southern Pashto`, `Spanish`, `Standard Latvian`, `Standard Malay`, `Swahili`, `Swedish`, `Tagalog`, `Tajik`, `Tamil`, `Telugu`, `Thai`, `Turkish`, `Ukrainian`, `Urdu`, `Vietnamese`, `Welsh`, `West Central Oromo`, `Western Persian`, `Yoruba`, `Zulu` | Language of input_text for T2ST and T2TT |
| `target_language_with_speech` | string | No | `French` | `Bengali`, `Catalan`, `Czech`, `Danish`, `Dutch`, `English`, `Estonian`, `Finnish`, `French`, `German`, `Hindi`, `Indonesian`, `Italian`, `Japanese`, `Korean`, `Maltese`, `Mandarin Chinese`, `Modern Standard Arabic`, `Northern Uzbek`, `Polish`, `Portuguese`, `Romanian`, `Russian`, `Slovak`, `Spanish`, `Swahili`, `Swedish`, `Tagalog`, `Telugu`, `Thai`, `Turkish`, `Ukrainian`, `Urdu`, `Vietnamese`, `Welsh`, `Western Persian` | Target language for tasks with speech output: S2ST and T2ST. Fewer languages are available for speech than for text output. |
| `target_language_text_only` | string | No | `Norwegian Nynorsk` | `Afrikaans`, `Amharic`, `Armenian`, `Assamese`, `Basque`, `Belarusian`, `Bengali`, `Bosnian`, `Bulgarian`, `Burmese`, `Cantonese`, `Catalan`, `Cebuano`, `Central Kurdish`, `Croatian`, `Czech`, `Danish`, `Dutch`, `Egyptian Arabic`, `English`, `Estonian`, `Finnish`, `French`, `Galician`, `Ganda`, `Georgian`, `German`, `Greek`, `Gujarati`, `Halh Mongolian`, `Hebrew`, `Hindi`, `Hungarian`, `Icelandic`, `Igbo`, `Indonesian`, `Irish`, `Italian`, `Japanese`, `Javanese`, `Kannada`, `Kazakh`, `Khmer`, `Korean`, `Kyrgyz`, `Lao`, `Lithuanian`, `Luo`, `Macedonian`, `Maithili`, `Malayalam`, `Maltese`, `Mandarin Chinese`, `Marathi`, `Meitei`, `Modern Standard Arabic`, `Moroccan Arabic`, `Nepali`, `North Azerbaijani`, `Northern Uzbek`, `Norwegian Bokmål`, `Norwegian Nynorsk`, `Nyanja`, `Odia`, `Polish`, `Portuguese`, `Punjabi`, `Romanian`, `Russian`, `Serbian`, `Shona`, `Sindhi`, `Slovak`, `Slovenian`, `Somali`, `Southern Pashto`, `Spanish`, `Standard Latvian`, `Standard Malay`, `Swahili`, `Swedish`, `Tagalog`, `Tajik`, `Tamil`, `Telugu`, `Thai`, `Turkish`, `Ukrainian`, `Urdu`, `Vietnamese`, `Welsh`, `West Central Oromo`, `Western Persian`, `Yoruba`, `Zulu` | Target language for tasks with text output only: S2TT, T2TT and ASR. |
| `max_input_audio_length` | number | No | `60` |  | Maximum input audio length |

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.0065 per API unit.
- **1 API unit = 1 request with up to 6 seconds of processing time.** A request that finishes within 6 seconds costs one unit ($0.0065).
- A request that runs longer bills one more unit for each started block of 6 seconds: more than 6 and up to 12 seconds is 2 units ($0.013), up to 18 seconds is 3 units ($0.0195), and so on.
- **Example cost**: the example request took 3.6 seconds of processing, which is 1 unit = **$0.0065**.
- Longer input audio or text takes longer and can use more units.
- **Failed or cancelled predictions are not billed.**
- **GET requests to check prediction status** are free and don't count toward API unit usage.
- **No monthly subscription fee.**

## API Documentation

### Base URL
All API requests go to: `https://prod.api.market/api/v1/magicapi/multilingual-multimodal-translation-cjwbw-seamless-communication`

### Authentication
Send your api.market API key in the `x-api-market-key` header:
```
x-api-market-key: YOUR_API_MARKET_KEY
```

### Asynchronous Processing
**IMPORTANT**: This API is asynchronous. The flow is:
1. Send a POST request to `/predictions`. A successful request returns **201 Created** with a prediction `id` and a `status` of `starting` or `processing`.
2. Poll GET `/predictions/{id}` with that ID.
3. Keep polling while `status` is `starting` or `processing`.
4. Stop when `status` is `succeeded` (read `output`), `failed` or `canceled` (read `error`).

A polling interval of 1 second and an overall timeout of 600 seconds (10 minutes) work well.

### Endpoints

#### Create a Prediction
```bash
curl -X POST "https://prod.api.market/api/v1/magicapi/multilingual-multimodal-translation-cjwbw-seamless-communication/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "version": "668a4fec05a887143e5fe8d45df25ec4c794dd43169b9a11562309b2d45873b0",
    "input": {
      "task_name": "S2ST (Speech to Speech translation)",
      "input_audio": "https://blog.api.market/wp-content/uploads/2026/10/seamless_communication-input-api-market.mp3",
      "input_text_language": "None",
      "max_input_audio_length": 60,
      "target_language_text_only": "Norwegian Nynorsk",
      "target_language_with_speech": "French"
    }
  }'
```

Response (201 Created):
```json
{
  "id": "21w3z6vbjhrmr0d1d0t889hwn8",
  "version": "668a4fec05a887143e5fe8d45df25ec4c794dd43169b9a11562309b2d45873b0",
  "status": "starting",
  "created_at": "2026-10-23T10:00:28.501Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/multilingual-multimodal-translation-cjwbw-seamless-communication/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK):
```json
{
  "id": "21w3z6vbjhrmr0d1d0t889hwn8",
  "status": "succeeded",
  "output": {
    "audio_output": "https://blog.api.market/wp-content/uploads/2026/10/seamless_communication-output-21w3z6vb-api-market.wav",
    "text_output": "Le modèle M4T sans faille de MetaAI démocratise la communication parlée à travers les barrières linguistiques."
  },
  "error": null,
  "created_at": "2026-10-23T10:00:28.501Z",
  "started_at": "2026-10-23T10:01:32.232Z",
  "completed_at": "2026-10-23T10:01:35.798Z"
}
```

`output` is an object. `text_output` (always present) holds the translated or transcribed text. `audio_output` is the URL of a WAV file for the speech-output tasks S2ST and T2ST.

### Status Codes
- **201 Created**: prediction created (POST `/predictions`). This is the normal success response.
- **200 OK**: prediction details returned (GET `/predictions/{id}`).
- **400 Bad Request**: invalid input, for example a `task_name` or target language that is not in the list.
- **401 Unauthorized**: missing or invalid API key.
- **404 Not Found**: no prediction with that ID.
- **429 Too Many Requests**: rate limit exceeded.
- **500 Internal Server Error** or **503 Service Unavailable**: temporary server problem; retry later.

## Rate Limits
All plans allow 10 requests per second. Requests above that return `429 Too Many Requests`; back off before retrying and honour any `Retry-After` header. Polling a prediction with GET is free but still counts toward the per-second limit, so poll every 1 to 5 seconds rather than in a tight loop.

## Error Handling
- Check the HTTP status code of every request and read the response body for the error message.
- A prediction can also end with `status` set to `failed`; the `error` field then explains why. Failed predictions are not billed.
- Retry 429, 500 and 503 responses with exponential backoff. Fix the input before retrying a 400.

## Implementation Guide
### Example Code

#### JavaScript
```javascript
const apiKey = "YOUR_API_MARKET_KEY";
const baseUrl = "https://prod.api.market/api/v1/magicapi/multilingual-multimodal-translation-cjwbw-seamless-communication";

async function createAndWaitForPrediction(input) {
  const response = await fetch(`${baseUrl}/predictions`, {
    method: "POST",
    headers: {
      "Content-Type": "application/json",
      "x-api-market-key": apiKey
    },
    body: JSON.stringify({ version: "668a4fec05a887143e5fe8d45df25ec4c794dd43169b9a11562309b2d45873b0", input })
  });

  if (response.status !== 201) {
    throw new Error(`API error: ${response.status} ${await response.text()}`);
  }

  const prediction = await response.json();
  return pollForResults(prediction.id);
}

const delay = (ms) => new Promise((resolve) => setTimeout(resolve, ms));

async function pollForResults(predictionId) {
  const maxTimeout = 600000;
  const pollInterval = 1000;
  const startTime = Date.now();

  while (Date.now() - startTime < maxTimeout) {
    const response = await fetch(`${baseUrl}/predictions/${predictionId}`, {
      headers: { "x-api-market-key": apiKey }
    });

    if (!response.ok) {
      throw new Error(`API error: ${response.status} ${await response.text()}`);
    }

    const prediction = await response.json();

    if (prediction.status === "succeeded") {
      return prediction.output;
    }

    if (prediction.status === "failed" || prediction.status === "canceled") {
      throw new Error(`Prediction ${prediction.status}: ${prediction.error || "Unknown error"}`);
    }

    await delay(pollInterval);
  }

  throw new Error("Prediction timed out after 10 minutes");
}

createAndWaitForPrediction({
  "task_name": "S2ST (Speech to Speech translation)",
  "input_audio": "https://blog.api.market/wp-content/uploads/2026/10/seamless_communication-input-api-market.mp3",
  "input_text_language": "None",
  "max_input_audio_length": 60,
  "target_language_text_only": "Norwegian Nynorsk",
  "target_language_with_speech": "French"
}).then((output) => console.log(output));
```

#### Python
```python
import time
import requests

API_KEY = "YOUR_API_MARKET_KEY"
BASE_URL = "https://prod.api.market/api/v1/magicapi/multilingual-multimodal-translation-cjwbw-seamless-communication"


def create_and_wait_for_prediction(input_data):
    headers = {"Content-Type": "application/json", "x-api-market-key": API_KEY}
    response = requests.post(
        f"{BASE_URL}/predictions",
        headers=headers,
        json={"version": "668a4fec05a887143e5fe8d45df25ec4c794dd43169b9a11562309b2d45873b0", "input": input_data},
    )

    if response.status_code != 201:
        raise Exception(f"API error: {response.status_code} {response.text}")

    return poll_for_results(response.json()["id"])


def poll_for_results(prediction_id):
    max_timeout = 600
    poll_interval = 1
    headers = {"x-api-market-key": API_KEY}
    start_time = time.time()

    while time.time() - start_time < max_timeout:
        response = requests.get(f"{BASE_URL}/predictions/{prediction_id}", headers=headers)

        if response.status_code != 200:
            raise Exception(f"API error: {response.status_code} {response.text}")

        prediction = response.json()

        if prediction["status"] == "succeeded":
            return prediction["output"]

        if prediction["status"] in ("failed", "canceled"):
            raise Exception(f"Prediction {prediction['status']}: {prediction.get('error') or 'Unknown error'}")

        time.sleep(poll_interval)

    raise Exception("Prediction timed out after 10 minutes")


output = create_and_wait_for_prediction({
    "task_name": "S2ST (Speech to Speech translation)",
    "input_audio": "https://blog.api.market/wp-content/uploads/2026/10/seamless_communication-input-api-market.mp3",
    "input_text_language": "None",
    "max_input_audio_length": 60,
    "target_language_text_only": "Norwegian Nynorsk",
    "target_language_with_speech": "French"
})
print(output)
```

## FAQ
- **Which fields do I send for each task?**
  - S2ST, S2TT and ASR need `input_audio`. T2ST and T2TT need `input_text` and `input_text_language`. S2ST and T2ST use `target_language_with_speech`; S2TT, T2TT and ASR use `target_language_text_only`.
- **Why are there fewer speech languages?**
  - Speech output is available in 36 languages, fewer than text output. Check the `target_language_with_speech` list.
- **Do speech-output tasks also return text?**
  - Yes. `text_output` holds the translated text and `audio_output` the WAV file.
- **How long can the input audio be?**
  - Set the cap with `max_input_audio_length` (default 60).
- **What did the example request cost?**
  - 1 API unit, $0.0065, because it took 3.6 seconds of processing.
- **Is the 201 response from POST an error?**
  - No. 201 Created means the prediction was accepted. Use the returned `id` to poll GET `/predictions/{id}` until `status` is `succeeded`, `failed` or `canceled`.

## Contact Us
If you have any questions, reach out to us through the chat on the bottom right of the page. You can also email our support team at hello@magicapi.com.


Translate between speech and text in dozens of languages, or transcribe speech, with Meta's SeamlessM4T in one API. $0.0065 per API unit, pay as you go.