Speech to Text

Transcribe an audio file into text, for example for voice input in a chatbot.

POST https://api.deepseek.com/v1/tools/asr

The whole audio file is sent in one request and the complete transcript is returned. Streaming recognition is not supported.

---

Languages

Chinese (zh), Cantonese (yue), English (en), Japanese (ja), Korean (ko), German (de), French (fr), Spanish (es), Portuguese (pt), Italian (it), Russian (ru), Arabic (ar), Hindi (hi), Indonesian (id), Vietnamese (vi), Thai (th), Turkish (tr), Malay (ms), Filipino (fil), Polish (pl) and Hungarian (hu).

There is no language parameter: the language is detected from the audio.

---

Request

Send the audio in the file field of a multipart/form-data request:

• Format: 16-bit PCM WAV or Ogg Opus, detected from the file content. WAV supports sample rates of 8000, 16000, 22050, 24000, 32000, 44100 and 48000 Hz.
• Channels: mono or stereo. For stereo audio, only the left channel is recognized.
• Size: at most 8 MiB. The upload must finish within 10 minutes.
• Duration: at most 120 seconds.

curl https://api.deepseek.com/v1/tools/asr \
  -H "Authorization: Bearer <DeepSeek API Key>" \
  -F file="@speech.wav"

---

Response

{
  "request_id": "...",
  "text": "What is the weather like in Shanghai today?",
  "duration_ms": 3200
}

Field	Description
request_id	The request ID
text	The transcript
duration_ms	The duration of the audio, in milliseconds

---

Errors

Status	Cause
400	The audio is not 16-bit PCM WAV or Ogg Opus, or the file field is missing
408	The upload did not finish within 10 minutes
413	The file is larger than 8 MiB
429	Your account has too many requests in progress
503	The service is busy. Retry later
504	Recognition timed out

---

Billing

Billing unit	List price	Limited-time price (50% off)
Per hour of audio	$0.116	$0.058

Billed by audio duration to the second; a fraction of a second counts as 1 second.

---

Rate Limits

An account can have at most 50 requests in progress; beyond that a request returns 429.