Tagalog speech-to-text API: 13% word error, measured

Measured accuracy

Model	Word error rate	Character error rate	Clips with no errors	Empty transcripts	Speed
Whisper Large v3 `openai/whisper-large-v3`	13%	4.4%	0 of 30	0 of 30	0.058× real time

Good. Usable with light review. Expect to correct a word here and there, mostly names and numbers. The clips are clean, single-speaker read speech, so treat this as a best case for your own recordings. How we measured, and all 83 languages.

What a transcript looks like

The clip with the median error rate from the test, reference text against what the API returned.

Spoken	Gayundin naman, sa pagkakaroon ng Schengen visa, hindi mo kailangang mag-apply ng visa nang hiwalay sa bawat bansang miyembro ng Schengen, kaya nakakatipid ng oras, pera at asikasuhing papeles.
Whisper Large v3	Gayun din naman, sa pagkakaroon ng Schengen Visa, hindi mo kailangang mag-apply ng visa ng hiwalay sa bawat bansang miyembro ng Schengen, kaya nakakatipid ng oras, pera, at asikasuhing papeles.

Language detection

All 10 clips we sent without a language field were identified as Tagalog. Auto-detection is safe here; sending language=tl still skips the detection step.

Quickstart

The request is the OpenAI transcription request with a different base URL. language is optional.

cURL

curl https://api.inferenceapis.com/v1/audio/transcriptions \
  -H "Authorization: Bearer $INFERENCE_API_KEY" \
  -F "model=openai/whisper-large-v3" \
  -F "language=tl" \
  -F "file=@recording.mp3" \
  -F "response_format=verbose_json"

Python · openai SDK

import os
from openai import OpenAI

client = OpenAI(base_url="https://api.inferenceapis.com/v1", api_key=os.environ["INFERENCE_API_KEY"])
with open("recording.mp3", "rb") as f:
    r = client.audio.transcriptions.create(model="openai/whisper-large-v3", file=f, language="tl", response_format="verbose_json")
print(r.text)

Subtitles and timestamps

response_format=srt or vtt returns a subtitle file with cues of up to two lines, broken at sentence ends and pauses. verbose_json with timestamp_granularities[]=word returns a start and end time for every word if you want to build your own.

Cost

$0.002 per minute of audio on either model: $0.12 for an hour, $12 for a hundred hours. Duration is measured from the file you upload. There is no daily cap on minutes and no subscription; enough to run your own Tagalog test.

FAQ

How accurate is Whisper for Tagalog?

On 30 clean read-speech clips it made 13% word errors, about one word in 8. Usable with light review. Expect to correct a word here and there, mostly names and numbers. Real recordings with noise, accents or several speakers will score worse.

Does language auto-detection work for Tagalog?

In our test, yes: all 10 clips sent without a language were identified as Tagalog. Sending language=tl still saves the detection step on every request.

Can it translate Tagalog speech to English?

The API returns Tagalog text. For English, transcribe first and translate the text with a chat model on the same key.

Related languages: Indonesian (5.5%) · Malay (8.6%)