Name: QwenASR API: Speech to Text and Forced Alignment
Brand: magicapi
SKU: api-magicapi-speech-recognition-alignment-api-twangodev-qwenasr
Availability: InStock
Author: API Market

Shashank Agarwal

# 🗣️ Speech-to-Text and Forced Alignment API with Word Timestamps (twangodev QwenASR)

## Overview
QwenASR by twangodev serves Qwen speech recognition and alignment. In `transcribe` mode it turns speech in an audio file into text with a start and end time for every word. In `align` mode it takes a transcript you already have and aligns it to the audio.

Send an `audio` file URL. The response is a JSON object with the detected `language`, the full `text` and a `timestamps` list.

## Features
- **Two modes**: `transcribe` (default) to get text from audio, or `align` to time an existing transcript.
- **Word timestamps**: every word comes with `start_time` and `end_time` in seconds.
- **11 languages for alignment**: Chinese, English, Cantonese, French, German, Italian, Japanese, Korean, Portuguese, Russian and Spanish, set with `language` (default `English`).
- **Simple input**: an audio URL plus, for `align` mode, the `text` to align.

## Use Cases
1. **Subtitles and captions**: word timings map directly to caption cues.
2. **Karaoke-style highlighting**: highlight each word as it is spoken in a player.
3. **Aligning scripts to recordings**: use `align` mode to time a script, lyrics you own or a manual transcript against the audio.
4. **Searchable recordings**: jump to the exact second a word is spoken.

## Visual Examples
### Example Input
Audio: [Listen to the example input](https://blog.api.market/wp-content/uploads/2026/10/qwenasr-input-v2-api-market.wav) (a short English voice clip)

Settings: `mode` `transcribe`, `language` `English`, `text` empty.

### Example Output
A JSON object with `language`, `text` and 18 word timestamps. Shortened (first three words shown):
```json
{
  "language": "English",
  "text": "Hey there, my name is Tara, and I'm a speech generation model that can sound like a person.",
  "timestamps": [
    { "start_time": 0.72, "end_time": 0.96, "text": "Hey" },
    { "start_time": 0.96, "end_time": 1.2, "text": "there" },
    { "start_time": 1.2, "end_time": 1.36, "text": "my" }
  ]
}
```
The last timestamp, "person", ends at 6.16 seconds.

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `audio` | string (URL) | Yes |  |  | Audio file to process |
| `mode` | string | No | `transcribe` | `transcribe`, `align` | Operation mode |
| `text` | string | No | `""` |  | Text to align (required when `mode` is `align`) |
| `language` | string | No | `English` | `Chinese`, `English`, `Cantonese`, `French`, `German`, `Italian`, `Japanese`, `Korean`, `Portuguese`, `Russian`, `Spanish` | Language (used with `align` mode) |

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.028 per API unit.
- **1 API unit = 1 request with up to 112 seconds of processing.** A request that takes longer is billed one more unit for each started 112-second block (for example, 130 seconds of processing is 2 units, $0.056).
- Longer audio takes longer to process and can add units.
- The example request took 26.0 seconds of processing, so it was billed 1 unit ($0.028).
- **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/speech-recognition-alignment-api-twangodev-qwenasr`

### 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/speech-recognition-alignment-api-twangodev-qwenasr/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "version": "fe8c3ddf8ba4ba14c935e8cb8cb5ec0761026719de678b7fad8078b7d2232ce7",
    "input": {
      "mode": "transcribe",
      "text": "",
      "audio": "https://blog.api.market/wp-content/uploads/2026/10/qwenasr-input-v2-api-market.wav",
      "language": "English"
    }
  }'
```

Response (201 Created):
```json
{
  "id": "4vtfjwpmjsrg80d1d9k8r2cxmc",
  "version": "fe8c3ddf8ba4ba14c935e8cb8cb5ec0761026719de678b7fad8078b7d2232ce7",
  "status": "starting",
  "created_at": "2026-10-23T20:14:46.605Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/speech-recognition-alignment-api-twangodev-qwenasr/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK, `timestamps` shortened):
```json
{
  "id": "4vtfjwpmjsrg80d1d9k8r2cxmc",
  "status": "succeeded",
  "output": {
    "language": "English",
    "text": "Hey there, my name is Tara, and I'm a speech generation model that can sound like a person.",
    "timestamps": [
      { "start_time": 0.72, "end_time": 0.96, "text": "Hey" },
      { "start_time": 0.96, "end_time": 1.2, "text": "there" },
      { "start_time": 1.2, "end_time": 1.36, "text": "my" }
    ]
  },
  "error": null,
  "created_at": "2026-10-23T20:14:46.605Z",
  "started_at": "2026-10-23T20:14:47.605Z",
  "completed_at": "2026-10-23T20:15:13.605Z"
}
```

`output` is an object with `language`, the full transcript in `text`, and `timestamps`: one entry per word with `text`, `start_time` and `end_time` in seconds. Punctuation is not included in the word entries.

### 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 missing `audio` or a `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 (for example, `align` mode with an empty `text`, or an audio URL that cannot be downloaded). 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/speech-recognition-alignment-api-twangodev-qwenasr";

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: "fe8c3ddf8ba4ba14c935e8cb8cb5ec0761026719de678b7fad8078b7d2232ce7", 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({
  mode: "transcribe",
  text: "",
  audio: "https://blog.api.market/wp-content/uploads/2026/10/qwenasr-input-v2-api-market.wav",
  language: "English"
}).then((output) => {
  console.log(output.text);
  for (const word of output.timestamps) {
    console.log(`${word.start_time}-${word.end_time}s ${word.text}`);
  }
});
```

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

API_KEY = "YOUR_API_MARKET_KEY"
BASE_URL = "https://prod.api.market/api/v1/magicapi/speech-recognition-alignment-api-twangodev-qwenasr"


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": "fe8c3ddf8ba4ba14c935e8cb8cb5ec0761026719de678b7fad8078b7d2232ce7", "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({
    "mode": "transcribe",
    "text": "",
    "audio": "https://blog.api.market/wp-content/uploads/2026/10/qwenasr-input-v2-api-market.wav",
    "language": "English",
})
print(output["text"])
for word in output["timestamps"]:
    print(f"{word['start_time']}-{word['end_time']}s {word['text']}")
```

## FAQ
- **What is the difference between `transcribe` and `align`?**
  - `transcribe` listens to the audio and writes the text. `align` takes text you already have (in `text`) and returns when each word is spoken.
- **Do I need to set `language`?**
  - It is used with `align` mode; set it to the language of your text. The default is `English`.
- **What does the output contain?**
  - `language`, the full `text`, and `timestamps` with one entry per word (`text`, `start_time`, `end_time` in seconds).
- **How much does a request cost?**
  - $0.028 for each started 112 seconds of processing. The example clip took 26.0 seconds, so it cost 1 unit ($0.028).
- **What if my prediction fails?**
  - Failed predictions are not billed. Read the `error` field, adjust the input and try again.

## 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.


Transcribe audio with word-level timestamps or align an existing transcript to speech in 11 languages with twangodev QwenASR. Pay as you go, no monthly fee.