About
The gpt-4o-mini-transcribe model is a highly efficient speech-to-text solution designed to deliver accurate audio transcriptions while optimizing for speed and resource consumption. This model offers significant improvements in word error rate and language recognition, making it particularly effective in scenarios involving accents, noisy environments, and varying speech speeds. gpt-4o-mini-transcribe is ideal for applications that require quick and reliable transcription services.
gpt-4o-mini-transcribe has been pretrained on specialized audio-centric datasets, which include diverse and high-quality audio samples, ensuring a deep understanding of speech nuances. This model supports a substantial context window of 16,000 tokens, allowing it to process longer audio inputs effectively. With a maximum output of 2,000 tokens, gpt-4o-mini-transcribe can generate detailed and comprehensive transcriptions. The training process incorporates rigorous enhancement techniques, including supervised fine-tuning and reinforcement learning, to optimize performance and accuracy.
1. Calling the API
Setup your API Key
To begin using Infron, you first need to create an account and generate your API Key.
Once you have your API key, set it as an environment variable in your runtime environment. This allows your applications to securely access Infron’s API.
export INFRON_KEY="YOUR_API_KEY"Submit a request
For long-running requests, you can poll for results.
curl https://media.onerouter.pro/v1/audios/generations \
-H "Authorization: Bearer YOUR_API_KEY" \
-F "model=openai/gpt-4o-mini-transcribe/audio-to-text" \
-F "file=@./files/audio.mp3" \
-F "language=en" \
-F "response_format=json"Response with queue task id
The response will look like this:
{
"code": 200,
"message": "success",
"data": {
"task_id": "6c462ff033ed483fa8ccfc3e94f78580",
"object": "audio",
"model": "openai/gpt-4o-mini-transcribe/audio-to-text",
"status": "created",
"urls": {
"query": "https://media.onerouter.pro/v1/audios/tasks/6c462ff033ed483fa8ccfc3e94f78580"
},
"created_at": "2026-05-31 14:36:47.81"
}
}Fetch request status
You can fetch the status of a request to check if it is completed or still in progress.
curl https://media.onerouter.pro/v1/audios/tasks/{task_id} \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json"
# For example
curl https://media.onerouter.pro/v1/audios/tasks/6c462ff033ed483fa8ccfc3e94f78580 \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json"Response with queue task status query
{
"code": 200,
"message": "success",
"data": {
"task_id": "6c462ff033ed483fa8ccfc3e94f78580",
"object": "audio",
"model": "openai/gpt-4o-mini-transcribe/audio-to-text",
"status": "completed",
"fail_reason": "",
"submit_time": 1780238208,
"start_time": 1780238207,
"finish_time": 1780238209,
"outputs": [
"Today is a wonderful day to build with Infran."
],
"usage": {
"audio_tokens": 29,
"completion_tokens": 0,
"input_token_details": {
"audio_tokens": 29,
"cached_creation_tokens": 0,
"cached_tokens": 0,
"image_tokens": 0,
"text_tokens": 0
},
"input_tokens_details": {
"audio_tokens": 29,
"cached_creation_tokens": 0,
"cached_tokens": 0,
"image_tokens": 0,
"text_tokens": 0
},
"output_count": 1,
"prompt_tokens": 29,
"prompt_tokens_details": {
"audio_tokens": 29
},
"total_tokens": 29
},
"cost": {
"total_cost": 8.7e-05,
"cost_details": {
"prompt_cost": 0,
"completion_cost": 0,
"image_cost": 0,
"video_cost": 0,
"audio_cost": 8.7e-05,
"native_web_search_cost": 0,
"plugin_web_search_cost": 0,
"tools_cost": 0,
"prompt_cache_read_cost": 0,
"prompt_cache_write_cost": 0,
"prompt_cache_write_5_min": 0,
"prompt_cache_write_1_h": 0,
"reasoning_cost": 0,
"discount_rate": 1,
"is_byok": false,
"byok_cost": 0
}
},
"created_at": "2026-05-31 14:36:48"
}
}Possible statuses:
- created
- in_progress
- processing
- completed
- failed
created -> in_progress → processing → completed | failed
Poll every 1-2 seconds until status is "completed" or "failed"
Get the result (when status is "completed")
Once the request is completed, you can fetch the result. See the Output Schema for the expected result format.
{
"code": 200,
"message": "success",
"data": {
"task_id": "6c462ff033ed483fa8ccfc3e94f78580",
"object": "audio",
"model": "openai/gpt-4o-mini-transcribe/audio-to-text",
"status": "completed",
"fail_reason": "",
"submit_time": 1780238208,
"start_time": 1780238207,
"finish_time": 1780238209,
"outputs": [
"Today is a wonderful day to build with Infran."
],
"usage": {
"audio_tokens": 29,
"completion_tokens": 0,
"input_token_details": {
"audio_tokens": 29,
"cached_creation_tokens": 0,
"cached_tokens": 0,
"image_tokens": 0,
"text_tokens": 0
},
"input_tokens_details": {
"audio_tokens": 29,
"cached_creation_tokens": 0,
"cached_tokens": 0,
"image_tokens": 0,
"text_tokens": 0
},
"output_count": 1,
"prompt_tokens": 29,
"prompt_tokens_details": {
"audio_tokens": 29
},
"total_tokens": 29
},
"cost": {
"total_cost": 8.7e-05,
"cost_details": {
"prompt_cost": 0,
"completion_cost": 0,
"image_cost": 0,
"video_cost": 0,
"audio_cost": 8.7e-05,
"native_web_search_cost": 0,
"plugin_web_search_cost": 0,
"tools_cost": 0,
"prompt_cache_read_cost": 0,
"prompt_cache_write_cost": 0,
"prompt_cache_write_5_min": 0,
"prompt_cache_write_1_h": 0,
"reasoning_cost": 0,
"discount_rate": 1,
"is_byok": false,
"byok_cost": 0
}
},
"created_at": "2026-05-31 14:36:48"
}
}2. Schema
Input
| Parameter | Type | Required | Default | Range | Description |
|---|---|---|---|---|---|
| model | string | Yes | -- | Explore all models | Model name. |
| file | file | Yes | -- | flac, mp3, mp4, mpeg, mpga, m4a, ogg, wav, webm | The audio file object (not file name) translate, in one of these formats: flac, mp3, mp4, mpeg, mpga, m4a, ogg, wav, or webm. |
| chunking_strategy | string | No | auto | -- | Controls how the audio is cut into chunks. When set to 'auto', the server first normalizes loudness and then uses voice activity detection (VAD) to choose boundaries. server_vad object can be provided to tweak VAD detection parameters manually. If unset, the audio is transcribed as a single block. |
| include | array | No | -- | -- | Additional information to include in the transcription response. logprobs will return the log probabilities of the tokens in the response to understand the model's confidence in the transcription. |
| knownspeakernames | array | No | -- | -- | Optional list of speaker names that correspond to the audio samples provided in knownspeakerreferences[]. Each entry should be a short identifier (for example customer oragent). Up to 4 speakers are supported. |
| knownspeakerreferences | array | No | -- | -- | Optional list of audio samples (as data URLs) that contain known speaker references matching knownspeakernames[]. Each sample must be between 2 and 10 seconds, and can use any of the same input audio formats supported by file. |
| language | string | No | en | ISO-639-1 (e.g. en) | The language of the input audio. Supplying the input language in ISO-639-1 (e.g. en) format will improve accuracy and latency. |
| prompt | string | No | -- | -- | An optional text to guide the model's style or continue a previous audio segment. The prompt should match the audio language. |
| response_format | string | No | json | json, text, srt, verbose_json, vtt, diarized_json | The format of the output. |
| temperature | number | No | 0 | >=0 and <=1 | The sampling temperature, between 0 and 1. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. If set to 0, the model will use log probability to automatically increase the temperature until certain thresholds are hit. |
Example request
curl https://media.onerouter.pro/v1/audios/generations \
-H "Authorization: Bearer YOUR_API_KEY" \
-F "model=openai/gpt-4o-mini-transcribe/audio-to-text" \
-F "file=@./files/audio.mp3" \
-F "language=en" \
-F "response_format=json"Output
| Parameter | Type | Range | Description |
|---|---|---|---|
| code | integer | -- | Status code of the request. |
| message | string | -- | Status message of the request. |
| data | object | -- | Data of the task. |
| data.task_id | string | -- | ID of the task. |
| data.object | string | -- | Object type of the task. |
| data.model | string | -- | Model name of the task. |
| data.status | string | created, in_progress, processing, completed, failed | Status of the task. |
| data.fail_reason | string | -- | Fail reason of the task. |
| data.submit_time | integer | -- | Timestamp of the task submission. |
| data.start_time | integer | -- | Timestamp of the task start. |
| data.finish_time | integer | -- | Timestamp of the task finish. |
| data.outputs | array | -- | Outputs of the task. |
| data.created_at | string | -- | Timestamp of the task creation. |
Example output
{
"code": 200,
"message": "success",
"data": {
"task_id": "6c462ff033ed483fa8ccfc3e94f78580",
"object": "audio",
"model": "openai/gpt-4o-mini-transcribe/audio-to-text",
"status": "completed",
"fail_reason": "",
"submit_time": 1780238208,
"start_time": 1780238207,
"finish_time": 1780238209,
"outputs": [
"Today is a wonderful day to build with Infran."
],
"usage": {
"audio_tokens": 29,
"completion_tokens": 0,
"input_token_details": {
"audio_tokens": 29,
"cached_creation_tokens": 0,
"cached_tokens": 0,
"image_tokens": 0,
"text_tokens": 0
},
"input_tokens_details": {
"audio_tokens": 29,
"cached_creation_tokens": 0,
"cached_tokens": 0,
"image_tokens": 0,
"text_tokens": 0
},
"output_count": 1,
"prompt_tokens": 29,
"prompt_tokens_details": {
"audio_tokens": 29
},
"total_tokens": 29
},
"cost": {
"total_cost": 8.7e-05,
"cost_details": {
"prompt_cost": 0,
"completion_cost": 0,
"image_cost": 0,
"video_cost": 0,
"audio_cost": 8.7e-05,
"native_web_search_cost": 0,
"plugin_web_search_cost": 0,
"tools_cost": 0,
"prompt_cache_read_cost": 0,
"prompt_cache_write_cost": 0,
"prompt_cache_write_5_min": 0,
"prompt_cache_write_1_h": 0,
"reasoning_cost": 0,
"discount_rate": 1,
"is_byok": false,
"byok_cost": 0
}
},
"created_at": "2026-05-31 14:36:48"
}
}3. Detailed Pricing
| Pricing Type | Unit | Price |
|---|---|---|
| Input audio | 1M tokens | $1.25 |
| Output text | 1M tokens | $5.00 |