Transcribe prerecorded audio and video in the language being spoken. Set a supported ISO language code, or let the API detect the likely language for each speech segment and return it with the transcript.
Welcome to our international customer interview.
Gracias por invitarnos a compartir nuestra experiencia.
Use one Jobs v2 integration for known-language recordings and media where the language is not known before submission.
Pass a supported code such as en, es, fr, or ja in the language parameter when you already know the recording language.
Leave language out, or send an empty value, and the API reports the likely language and detection confidence for each speech segment.
The transcript stays in the language being spoken. Transcription does not translate the recording into English or another language.
Completed JSON segments keep their timestamps, text, and language value together, ready for routing, indexing, review, or downstream processing.
API Example
This Jobs v2 request enables speaker diarization, word timestamps, and webhook delivery without setting a language. The completed output reports language metadata on each segment.
curl --request POST \
--url https://api.speechischeap.com/v2/jobs/ \
--header 'Authorization: Bearer YOUR_API_KEY' \
--header 'Content-Type: application/json' \
--data '{
"input_url": "https://example.com/international-interview.mp3",
"can_parse_speakers": true,
"can_parse_words": true,
"webhook_url": "https://your-domain.com/transcription-webhook"
}'{
"id": "00000000-1111-7222-b333-444444444444-sic",
"status": "COMPLETED",
"output": {
"segments": [
{
"id": 1,
"start": 0.0,
"end": 4.86,
"speaker_id": "A",
"language": "en (99.95%)",
"confidence": 0.987,
"text": "Welcome to our international customer interview."
},
{
"id": 2,
"start": 5.14,
"end": 9.72,
"speaker_id": "B",
"language": "es (99.72%)",
"confidence": 0.976,
"text": "Gracias por invitarnos a compartir nuestra experiencia."
}
]
}
}Already Know the Language?
Set the ISO code directly. This compact request tells Jobs v2 to transcribe the recording as French.
{
"input_url": "https://example.com/french-podcast.mp3",
"language": "fr",
"can_parse_words": true
}Supported Languages
Filter by language name or ISO code. This list matches the canonical supported-language documentation.
Showing all 100 languages
afsqamarhyasazbaeubebnbsbrbgyuecazhhrcsdanlenetfofifrglkadeelguhthahawhehihuisiditjajwknkkkmkololalvlnltlbmkmgmsmlmtmimrmnmynenonnocpsfaplptparorusasrsnsdsiskslsoessuswsvtltgtatttethbotrtkukuruzvicyyiyoLanguage control works alongside the output and delivery options used by the rest of your transcription integration.
Add anonymous speaker labels to multilingual interviews, support calls, podcasts, and meetings with can_parse_speakers.
Set can_parse_words to receive word-level timing data for subtitles, review tools, editing, and search interfaces.
Submit prerecorded media asynchronously, then poll by job ID or receive completed output at your webhook URL.
Keep segment timestamps, language metadata, confidence, text, and enabled add-on fields in one machine-readable result.
Use one integration for prerecorded speech collected across countries, teams, customers, and media libraries.
Accept customer audio and video in supported languages without building a separate transcription integration for every market.
Turn prerecorded support conversations into searchable text while keeping each segment in the language that was spoken.
Add transcript text and language metadata to audio and video libraries so teams can route and index each recording.
Transcribe complete episodes and interviews, with optional speaker labels and word timestamps for publishing workflows.
Process collections of prerecorded speech into structured segments for review, annotation, indexing, or analysis.
Create same-language transcripts for recorded presentations, lessons, events, and internal media across supported languages.
Simple Core Pricing
Core transcription pricing does not change when you set a language or use automatic detection. Optional add-ons are billed separately.
Pay as You Go
Start with no upfront cost and pay for successfully completed transcriptions.
1 hour: $0.12
10 hours: $1.20
Subscriber
Includes 21,600 core transcription minutes, equal to 15 days of audio.
Additional core transcription minutes are billed at $0.000926 each.
Clear Boundaries
The API returns transcript text in the language being spoken. Language detection helps identify speech segments, but it does not remove every language-specific source of error.
Spanish speech returns Spanish text. Translate it separately if your product needs another language.
Language, accent, audio quality, noise, vocabulary, and context can all affect the transcript.
Per-segment detection can report different languages, but switching within a sentence may not be identified perfectly.
Language selection, automatic detection, output, and integration details.
The current supported-language list contains 100 languages. Use the searchable list on this page or check the supported-language docs for the canonical list.
Add the supported ISO code to the language request parameter, such as "language": "fr" for French. Leave the parameter out, or send an empty value, to use automatic detection for each speech segment.
No. Speech is Cheap transcribes speech in the language being spoken. A Spanish recording returns a Spanish transcript. Translation is a separate step outside this transcription API.
Automatic detection reports a likely language for each speech segment, which can help with recordings that contain separate passages in different languages. It does not promise perfect switching inside a sentence, and results can vary by language, accent, recording quality, and context.
Yes. The same Jobs v2 request can enable can_parse_speakers, can_parse_words, and a webhook_url. Completed JSON output keeps language metadata alongside the segment text and optional speaker or word data.
Choose a plan, get your API key, and submit prerecorded audio or video with an explicit language or automatic detection.