Thai Speech-to-Text

Turns a recording of Thai speech into text, one timestamped segment per speaker turn. Two tiers take the same request and return the same response: Base answers in 0.3 s typical at 1 IC per minute; Pro trades speed for accuracy, 97.81% of characters correct, at 2 IC per minute.

POST/v3/store/speech/speech-to-text/base

+1 endpoints

Try it

• Price1 ICper 60 seconds
• Served317.2Kcalls

• v3.0
• Active

Input

OutputPOST /v3/store/speech/speech-to-text/base

output

2 × text: สวัสดีครับ วันนี้เรามาคุยเรื่องการใช้งาน API กันนะครับ start: 0 end: 4.6 speaker: SPEAKER_00 segment: 0

audio_duration_in_seconds

8.1

uploaded_file_name

recording.wav

processing_time_in_seconds

0.3

iapp

seconds: 8.1

Every field, under ResponseOther endpoints: /pro see Request

Request

curl -X POST https://api.iapp.co.th/v3/store/speech/speech-to-text/base \
  -H "apikey: YOUR_API_KEY" \
  -F "file=@recording.wav"

Response

200OK. The fields are explained under Response.

{
  "output": [
    { "text": "สวัสดีครับ วันนี้เรามาคุยเรื่องการใช้งาน API กันนะครับ", "start": 0.0, "end": 4.6, "speaker": "SPEAKER_00", "segment": 0 },
    { "text": "ได้เลยค่ะ เริ่มจากการขอ API key ก่อนนะคะ", "start": 4.6, "end": 8.1, "speaker": "SPEAKER_01", "segment": 1 }
  ],
  "audio_duration_in_seconds": 8.1,
  "uploaded_file_name": "recording.wav",
  "processing_time_in_seconds": 0.3,
  "iapp": { "seconds": 8.1 }
}

Request

multipart/form-data, one recording per request, to the endpoint of the tier you want.

Tier	Endpoint	Price	Character error rate
Base	POST /v3/store/speech/speech-to-text/base	1 IC per 60 seconds	5.87%
Pro	POST /v3/store/speech/speech-to-text/pro	2 IC per 60 seconds	2.19%

Headers

apikeystringrequired

Your API key, created under API Keys in the control panel.

Body

filefilerequired

The recording. MP3, WAV, AAC or M4A, up to 1 GB.

chunk_sizeinteger

Seconds of audio per processing chunk. Default 7.

The legacy paths /asr/v3 and /asr/v3/pro answer at the same prices.

Response

outputarray

The segments, in the order spoken

output.textstring

What was said

output.startnumber

When the segment starts, in seconds

output.endnumber

When it ends, in seconds

output.speakerstring

The speaker's label: SPEAKER_00, SPEAKER_01 and so on

output.segmentinteger

The segment's index, from 0

audio_duration_in_secondsprocessing_time_in_secondsuploaded_file_name

The recording's length, the time taken, the file name

iappobject

Billing

iapp.secondsnumber

The seconds billed

Status codes

Code	Meaning
400	No file, or an unsupported format
404	Task not found
500	Processing failed

Details

Measured

On the Common Voice 17.0 Thai test set: 11,042 samples the models had not seen, with men, women and children among the speakers.

Measure	Base	Pro
Word error rate	14.52%, so 85.48% of words correct	8.01%, so 91.99% of words correct
Character error rate	5.87%, so 94.13% of characters correct	2.19%, so 97.81% of characters correct
Response time, typical	0.3 s	slower than Base; the accuracy is what it buys

Per-sample results for Base: the benchmark sheet.

Limits

• Thai speech, up to 30 minutes of audio per request, on either tier.

Data handling

The service is GDPR and PDPA compliant.