{ "aboutHtml": "

GPT Transcribe is OpenAI’s speech-to-text model for completed audio files, streamed file transcripts and committed audio turns in Realtime sessions. It accepts audio with optional text context and returns recognized text. The producer positions it for accurate transcription of recorded speech in its original language, including domain vocabulary, multilingual recordings and code-switching. Its useful output is a transcript that stays connected to the recording, with detected-language metadata when the model can make a reliable prediction.

Preparing a completed recording

OpenAI’s file-transcription workflow accepts uploads up to 25 MB. The current endpoint reference lists FLAC, MP3, MP4, MPEG, MPGA, M4A, OGG, WAV and WebM input formats and recommends enough file metadata to identify the format, including an appropriate filename extension and content type. For a larger recording, the guide recommends compression or splitting it into files of 25 MB or less. Avoid cutting through a sentence, because removing surrounding speech can make recognition less accurate. Keep the pieces in order and retain their positions in the original recording.

For a meeting, interview or recorded presentation, supply the actual audio and identify the topic and any terminology likely to be spoken. A useful result is the recognized wording rather than an inferred summary of the discussion. Preserve the recording’s identifier with the returned text so that a person reviewing a name, number or uncertain phrase can return to the right source. A transcript can then become material for a separate summary, translation or editorial workflow.

Context, keyword hints and expected languages

The producer separates three kinds of guidance. A prompt gives free-form context about the recording’s topic or setting. Keywords give literal words or phrases expected in the speech. A languages list identifies the expected spoken languages. For a bilingual product discussion, describe the setting in the prompt, list the relevant product names as keywords, and provide the expected languages. These are recognition hints; they are not instructions to insert vocabulary that nobody said.

GPT Transcribe uses the plural languages field instead of the older singular language field; OpenAI says not to send both. Its guide accepts supported ISO 639-1 codes, selected ISO 639-3 codes and regional Chinese locale codes, with unsupported or malformed values rejected. Each keyword must stay on one line and must not contain an angle bracket, carriage return or line feed. An invalid keyword or an overlong prompt can reject the complete request. Keep context relevant and check whether hinted names actually occur in the audio rather than treating the hint list as the transcript.

JSON output and detected-language predictions

The documented file result is JSON containing the transcript and detected languages. Those language predictions are the model’s output, not a copy of the supplied language hints. An empty language list is a valid result when the model cannot make a reliable prediction; it does not establish that the recording was silent or that transcription failed. Keep the recognized text and language metadata separate so downstream processing can use each according to its meaning.

Streaming a file transcript

A completed file can produce transcript text incrementally while it is processed. OpenAI documents transcript.text.delta events for partial text and transcript.text.done for the complete file transcript; the final event also includes detected languages. This streams the output of an already supplied recording. It is useful for displaying progress on a long upload, while the final event establishes the text to retain as the completed result.

Committed turns in Realtime sessions

GPT Transcribe also has a specialized Realtime workflow over WebSocket. Audio is appended to the input buffer, then the application explicitly commits an audio turn. Transcription begins for that committed chunk and may emit deltas before its completion event, including detected-language information. The model automatically uses earlier transcribed turns as context in this producer session flow. This is distinct from continuously emitting live-caption text as microphone audio arrives; OpenAI names GPT Live Transcribe for that continuous workflow.

For several turns, retain the input-item identifier with the corresponding transcript. OpenAI warns that completion events from different turns need not arrive in speech order, so associate each final result with its committed input rather than arranging a conversation solely by completion time. The useful artifact is an ordered transcript assembled from the correct final items, with interim text treated as progress rather than a separate completed utterance.

Specialized outputs and substantive limits

This profile does not supply a universal subtitle, timestamp, translation or speaker-label contract. OpenAI’s migration guide distinguishes those specialized capabilities: native subtitle formats and word timing require a compatible model and format, while speaker labels use GPT-4o Transcribe Diarize. GPT Transcribe’s file result is JSON rather than a drop-in SRT, VTT or verbose-timestamp response. Its exact model card does not publish a numeric context or maximum-output allowance, so a chat-model token window should not be assumed for audio transcription.

Review difficult vocabulary, numbers, accents, background noise and changes of language against the recording. Check that relevant hints improved recognition without introducing unspoken terms and that the final transcript includes the end of the audio. For a multilingual result, preserve the spoken wording before preparing any translated version. These checks make the transcript useful as source material while keeping context hints, detected languages and the recognized speech clearly distinguished.

", "aboutText": "GPT Transcribe is OpenAI’s speech-to-text model for completed audio files, streamed file transcripts and committed audio turns in Realtime sessions. It accepts audio with optional text context and returns recognized text. The producer positions it for accurate transcription of recorded speech in its original language, including domain vocabulary, multilingual recordings and code-switching. Its useful output is a transcript that stays connected to the recording, with detected-language metadata when the model can make a reliable prediction.\n\n### Preparing a completed recording\n\nOpenAI’s file-transcription workflow accepts uploads up to 25 MB. The current endpoint reference lists FLAC, MP3, MP4, MPEG, MPGA, M4A, OGG, WAV and WebM input formats and recommends enough file metadata to identify the format, including an appropriate filename extension and content type. For a larger recording, the guide recommends compression or splitting it into files of 25 MB or less. Avoid cutting through a sentence, because removing surrounding speech can make recognition less accurate. Keep the pieces in order and retain their positions in the original recording.\n\nFor a meeting, interview or recorded presentation, supply the actual audio and identify the topic and any terminology likely to be spoken. A useful result is the recognized wording rather than an inferred summary of the discussion. Preserve the recording’s identifier with the returned text so that a person reviewing a name, number or uncertain phrase can return to the right source. A transcript can then become material for a separate summary, translation or editorial workflow.\n\n### Context, keyword hints and expected languages\n\nThe producer separates three kinds of guidance. A prompt gives free-form context about the recording’s topic or setting. Keywords give literal words or phrases expected in the speech. A languages list identifies the expected spoken languages. For a bilingual product discussion, describe the setting in the prompt, list the relevant product names as keywords, and provide the expected languages. These are recognition hints; they are not instructions to insert vocabulary that nobody said.\n\nGPT Transcribe uses the plural languages field instead of the older singular language field; OpenAI says not to send both. Its guide accepts supported ISO 639-1 codes, selected ISO 639-3 codes and regional Chinese locale codes, with unsupported or malformed values rejected. Each keyword must stay on one line and must not contain an angle bracket, carriage return or line feed. An invalid keyword or an overlong prompt can reject the complete request. Keep context relevant and check whether hinted names actually occur in the audio rather than treating the hint list as the transcript.\n\n### JSON output and detected-language predictions\n\nThe documented file result is JSON containing the transcript and detected languages. Those language predictions are the model’s output, not a copy of the supplied language hints. An empty language list is a valid result when the model cannot make a reliable prediction; it does not establish that the recording was silent or that transcription failed. Keep the recognized text and language metadata separate so downstream processing can use each according to its meaning.\n\n### Streaming a file transcript\n\nA completed file can produce transcript text incrementally while it is processed. OpenAI documents transcript.text.delta events for partial text and transcript.text.done for the complete file transcript; the final event also includes detected languages. This streams the output of an already supplied recording. It is useful for displaying progress on a long upload, while the final event establishes the text to retain as the completed result.\n\n### Committed turns in Realtime sessions\n\nGPT Transcribe also has a specialized Realtime workflow over WebSocket. Audio is appended to the input buffer, then the application explicitly commits an audio turn. Transcription begins for that committed chunk and may emit deltas before its completion event, including detected-language information. The model automatically uses earlier transcribed turns as context in this producer session flow. This is distinct from continuously emitting live-caption text as microphone audio arrives; OpenAI names GPT Live Transcribe for that continuous workflow.\n\nFor several turns, retain the input-item identifier with the corresponding transcript. OpenAI warns that completion events from different turns need not arrive in speech order, so associate each final result with its committed input rather than arranging a conversation solely by completion time. The useful artifact is an ordered transcript assembled from the correct final items, with interim text treated as progress rather than a separate completed utterance.\n\n### Specialized outputs and substantive limits\n\nThis profile does not supply a universal subtitle, timestamp, translation or speaker-label contract. OpenAI’s migration guide distinguishes those specialized capabilities: native subtitle formats and word timing require a compatible model and format, while speaker labels use GPT-4o Transcribe Diarize. GPT Transcribe’s file result is JSON rather than a drop-in SRT, VTT or verbose-timestamp response. Its exact model card does not publish a numeric context or maximum-output allowance, so a chat-model token window should not be assumed for audio transcription.\n\nReview difficult vocabulary, numbers, accents, background noise and changes of language against the recording. Check that relevant hints improved recognition without introducing unspoken terms and that the final transcript includes the end of the audio. For a multilingual result, preserve the spoken wording before preparing any translated version. These checks make the transcript useful as source material while keeping context hints, detected languages and the recognized speech clearly distinguished.", "editorialOverlay": { "copy_fields": null, "copy_sha256": null, "copy_source": null, "html_sha256": "9d346f99d47c188665b68ed40ffcbd4d903d92ebeb72aa22efcbf6c843eede37", "module_sha256": "ee9dc1c50b1e6a8fb7fa062030fcb4646b9aaaeb08be513b723a5a92a7290f06", "raw_model_unchanged": true, "schema": "ailoot-native-model-editorial-v1" }, "locale": "en", "methods": [ { "html": "
O

OpenAI: GPT Transcribe

OpenAI /openai/gpt-transcribe

Overview

GPT Transcribe turns recorded speech into text in its original language, with contextual prompts, literal keyword hints and multiple language hints. OpenAI documents completed-file transcription, streamed file results and transcription after committed Realtime audio turns.

Accepted inputs
Audio
Results
Text
About the model

This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.

GPT Transcribe is OpenAI’s speech-to-text model for completed audio files, streamed file transcripts and committed audio turns in Realtime sessions. It accepts audio with optional text context and returns recognized text. The producer positions it for accurate transcription of recorded speech in its original language, including domain vocabulary, multilingual recordings and code-switching. Its useful output is a transcript that stays connected to the recording, with detected-language metadata when the model can make a reliable prediction.

Preparing a completed recording

OpenAI’s file-transcription workflow accepts uploads up to 25 MB. The current endpoint reference lists FLAC, MP3, MP4, MPEG, MPGA, M4A, OGG, WAV and WebM input formats and recommends enough file metadata to identify the format, including an appropriate filename extension and content type. For a larger recording, the guide recommends compression or splitting it into files of 25 MB or less. Avoid cutting through a sentence, because removing surrounding speech can make recognition less accurate. Keep the pieces in order and retain their positions in the original recording.

For a meeting, interview or recorded presentation, supply the actual audio and identify the topic and any terminology likely to be spoken. A useful result is the recognized wording rather than an inferred summary of the discussion. Preserve the recording’s identifier with the returned text so that a person reviewing a name, number or uncertain phrase can return to the right source. A transcript can then become material for a separate summary, translation or editorial workflow.

Context, keyword hints and expected languages

The producer separates three kinds of guidance. A prompt gives free-form context about the recording’s topic or setting. Keywords give literal words or phrases expected in the speech. A languages list identifies the expected spoken languages. For a bilingual product discussion, describe the setting in the prompt, list the relevant product names as keywords, and provide the expected languages. These are recognition hints; they are not instructions to insert vocabulary that nobody said.

GPT Transcribe uses the plural languages field instead of the older singular language field; OpenAI says not to send both. Its guide accepts supported ISO 639-1 codes, selected ISO 639-3 codes and regional Chinese locale codes, with unsupported or malformed values rejected. Each keyword must stay on one line and must not contain an angle bracket, carriage return or line feed. An invalid keyword or an overlong prompt can reject the complete request. Keep context relevant and check whether hinted names actually occur in the audio rather than treating the hint list as the transcript.

JSON output and detected-language predictions

The documented file result is JSON containing the transcript and detected languages. Those language predictions are the model’s output, not a copy of the supplied language hints. An empty language list is a valid result when the model cannot make a reliable prediction; it does not establish that the recording was silent or that transcription failed. Keep the recognized text and language metadata separate so downstream processing can use each according to its meaning.

Streaming a file transcript

A completed file can produce transcript text incrementally while it is processed. OpenAI documents transcript.text.delta events for partial text and transcript.text.done for the complete file transcript; the final event also includes detected languages. This streams the output of an already supplied recording. It is useful for displaying progress on a long upload, while the final event establishes the text to retain as the completed result.

Committed turns in Realtime sessions

GPT Transcribe also has a specialized Realtime workflow over WebSocket. Audio is appended to the input buffer, then the application explicitly commits an audio turn. Transcription begins for that committed chunk and may emit deltas before its completion event, including detected-language information. The model automatically uses earlier transcribed turns as context in this producer session flow. This is distinct from continuously emitting live-caption text as microphone audio arrives; OpenAI names GPT Live Transcribe for that continuous workflow.

For several turns, retain the input-item identifier with the corresponding transcript. OpenAI warns that completion events from different turns need not arrive in speech order, so associate each final result with its committed input rather than arranging a conversation solely by completion time. The useful artifact is an ordered transcript assembled from the correct final items, with interim text treated as progress rather than a separate completed utterance.

Specialized outputs and substantive limits

This profile does not supply a universal subtitle, timestamp, translation or speaker-label contract. OpenAI’s migration guide distinguishes those specialized capabilities: native subtitle formats and word timing require a compatible model and format, while speaker labels use GPT-4o Transcribe Diarize. GPT Transcribe’s file result is JSON rather than a drop-in SRT, VTT or verbose-timestamp response. Its exact model card does not publish a numeric context or maximum-output allowance, so a chat-model token window should not be assumed for audio transcription.

Review difficult vocabulary, numbers, accents, background noise and changes of language against the recording. Check that relevant hints improved recognition without introducing unspoken terms and that the final transcript includes the end of the audio. For a multilingual result, preserve the spoken wording before preparing any translated version. These checks make the transcript useful as source material while keeping context hints, detected languages and the recognized speech clearly distinguished.

Available in Ailoot · inputs & methods

Ailoot Ultra · Transcription POST /v1/audio/transcriptions
Input types
audio
Output types
text
FieldTypeRequirementValues & limitsDefault
FieldmodelTypestringRequirementRequiredValues & limits
openai/gpt-transcribe
Default—
Fieldinput_audioTypeobjectRequirementRequiredValues & limitsDefault—
Method parameters and lifecycle
ParameterContract
Parametercontent_typeContractapplication/json
ParameterrequiredContractmodel, input_audio
Parametermaterial_typesContractaudio
Parametermaterial_conditionsContractThe example reads your actual nonempty PCM WAV file, encodes the complete file bytes as base64 and sends input_audio.format as wav.

Lifecycle: synchronous_result_then_owned_library

GET /v1/media/tasks/{task_id} · exact task_id from an existing owned media record

GET /v1/media/assets/{asset_id}/content

GET /v1/media/assets

GET /v1/media/assets/{asset_id}

Pricing

Tariff references do not add request options. Supported parameters are listed for the selected method.

Fast mode
Not specified When fast mode is selected
Flex mode
Not specified When flex mode is selected
Maximum reasoning effort
Not specified When maximum reasoning effort uses the default adjustment

Ailoot rates

Prices by configuration

Match each charge to the conditions shown. Use one rate for each set of alternative configurations.

ConfigurationRate scheduleBilling unitCharge
ConfigurationBase token ratesRate scheduleBase rateBilling unit—Charge
Input · Input tokens — / 1K tokens
Output · Output tokens — / 1K tokens
Input · Cache read — / 1K tokens
Input · Cache write — / 1K tokens
Input · Cache write, 1 hour — / 1K tokens
Configurationinput audio secondRate schedule—Billing unitper input audio secondCharge
Input · input audio second $0.00015 per input audio second

Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →

Example requests

Ailoot Ultra · Transcription

set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\n: "${AILOOT_AUDIO_FILE:?Set the path to your actual PCM WAV recording}"\nexport AILOOT_AUDIO_FILE\npython3 - <<'PY_REQUEST'\nimport base64, io, json, os, wave\nfrom pathlib import Path\naudio = Path(os.environ['AILOOT_AUDIO_FILE']).read_bytes()\nif not (audio[:4] == b'RIFF' and audio[8:12] == b'WAVE'):\n    raise SystemExit('Supply your actual PCM WAV audio file')\nwith wave.open(io.BytesIO(audio), 'rb') as recording:\n    if recording.getcomptype() != 'NONE' or recording.getnframes() <= 0:\n        raise SystemExit('Supply a nonempty PCM WAV recording')\nrequest = {'model': 'openai/gpt-transcribe', 'input_audio': {'data': base64.b64encode(audio).decode('ascii'), 'format': 'wav'}}\nPath('request.json').write_text(json.dumps(request, ensure_ascii=False), encoding='utf-8')\nPY_REQUEST\ncurl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/audio/transcriptions \\\n  -H "Authorization: Bearer $AILOOT_API_KEY" -H 'Content-Type: application/json' \\\n  --data-binary @request.json -D response-headers.txt -o response.bin\n# Retain the complete original response, headers and actual X-Client-Request-ID.\n# Read existing owned records for recovery; do not resubmit an uncertain operation.\n

Receive and save results

Ailoot Ultra · Transcription

Complete original response body, status and forwarded headers. No universal JSON shape, text field or task envelope is added by this handler.

Keep all completed results

Keep every complete returned response, file and content part.

Retain the complete response and headers, every actual output, original returned identifiers and all usage fields. Recover the original operation without submitting creation again.

If the original response needs checking

Retain the actual X-Client-Request-ID and complete response. Read the original owned assets and their exact signed content URLs; never resubmit an uncertain creation.

Response fields
FieldType
Fieldtext (when present)Type
Fieldusage (when present)Type
Errors and next steps
ResponseConditionNext step
Response400 · invalid_request_errorConditionThe JSON body or explicit canonical model ID is invalid.Next stepRetain the response and correct the indicated request field.
Response401 · authentication_errorConditionThe Ailoot API key is invalid.Next stepUse your authorized Ailoot API key.
Response403 · permission_errorConditionThe key group does not admit this method.Next stepRetain the returned error details.
Response404 · not_found_errorConditionThe selected method has no admitted route.Next stepRetain the returned error details.
Response422 · price_unavailableConditionThe accepted tariff cannot obtain its required input quantity or bound.Next stepRetain the response; use the input representation required by the accepted tariff.
Response503 · ConditionRequired storage, route, method price or reservation is unavailable.Next stepKeep the complete original response. An uncertain reservation is not permission to resubmit.
Response502 · server_errorConditionThe original result, usage or settlement requires checking.Next stepRetain the actual request ID and original response. Read the existing owned record; do not resubmit creation.

Aliases

Model aliases

Ailoot Ultra openai/gpt-transcribe

Method path aliases

Use the method paths shown above.

Related caller IDs

openai/gpt-transcribe

", "method_identity": { "effective_request_model_id": "openai/gpt-transcribe", "fmt": "transcriptions", "http": "POST", "path": "/v1/audio/transcriptions", "request_model_id": "openai/gpt-transcribe", "route_label": "Ultra" }, "method_key": "Ultra|transcriptions|/v1/audio/transcriptions|openai%2Fgpt-transcribe", "ordinal": 0, "rawMethod": { "available": null, "cache_read": null, "cache_write": null, "cache_write_1h": null, "contract": { "data": { "all_results": "Retain the complete response and headers, every actual output, original returned identifiers and all usage fields. Recover the original operation without submitting creation again.", "errors": [ { "code": null, "condition": "The JSON body or explicit canonical model ID is invalid.", "http_status": 400, "recovery": "Retain the response and correct the indicated request field.", "type": "invalid_request_error" }, { "code": null, "condition": "The Ailoot API key is invalid.", "http_status": 401, "recovery": "Use your authorized Ailoot API key.", "type": "authentication_error" }, { "code": null, "condition": "The key group does not admit this method.", "http_status": 403, "recovery": "Retain the returned error details.", "type": "permission_error" }, { "code": null, "condition": "The selected method has no admitted route.", "http_status": 404, "recovery": "Retain the returned error details.", "type": "not_found_error" }, { "code": null, "condition": "The accepted tariff cannot obtain its required input quantity or bound.", "http_status": 422, "recovery": "Retain the response; use the input representation required by the accepted tariff.", "type": "price_unavailable" }, { "code": null, "condition": "Required storage, route, method price or reservation is unavailable.", "http_status": 503, "recovery": "Keep the complete original response. An uncertain reservation is not permission to resubmit.", "type": null }, { "code": null, "condition": "The original result, usage or settlement requires checking.", "http_status": 502, "operation_id_source": "actual X-Client-Request-ID", "recovery": "Retain the actual request ID and original response. Read the existing owned record; do not resubmit creation.", "response_headers": { "x-should-retry": "false" }, "type": "server_error" } ], "lifecycle": { "assets": [ { "http": "GET", "path": "/v1/media/assets" }, { "http": "GET", "path": "/v1/media/assets/{asset_id}" } ], "content": [ { "http": "GET", "path": "/v1/media/assets/{asset_id}/content", "url_source": "exact returned signed content URL" } ], "dedicated_method_polling_or_cancel": false, "mode": "synchronous_result_then_owned_library", "original_POST_adds_task_envelope": false, "status": [ { "http": "GET", "id_source": "exact task_id from an existing owned media record", "path": "/v1/media/tasks/{task_id}" } ], "unknown": "Retain the actual X-Client-Request-ID and complete response. Read the original owned assets and their exact signed content URLs; never resubmit an uncertain creation." }, "outputs": { "all_results": "Retain the complete response and headers, every actual output, original returned identifiers and all usage fields. Recover the original operation without submitting creation again.", "family": "text", "fields": [ { "description": null, "name": "text (when present)", "type": null }, { "description": null, "name": "usage (when present)", "type": null } ], "result": "Complete original response body, status and forwarded headers. No universal JSON shape, text field or task envelope is added by this handler." }, "parameters": [ { "default_known": false, "enum": [ "openai/gpt-transcribe" ], "enum_known": true, "name": "model", "required": true, "type": "string" }, { "default_known": false, "enum_known": false, "name": "input_audio", "required": true, "type": "object" } ], "request_schema": { "properties": { "input_audio": { "properties": { "data": { "contentEncoding": "base64", "type": "string" }, "format": { "type": "string" } }, "required": [ "data", "format" ], "type": "object" }, "model": { "enum": [ "openai/gpt-transcribe" ], "type": "string" } }, "required": [ "model", "input_audio" ], "type": "object" }, "schema_template": { "actual_file_preparation_required": true, "body": { "input_audio": { "data": "${BASE64_ACTUAL_WAV_BYTES}", "format": "wav" }, "model": "openai/gpt-transcribe" }, "content_type": "application/json" } }, "input": { "content_type": "application/json", "material_conditions": [ "The example reads your actual nonempty PCM WAV file, encodes the complete file bytes as base64 and sends input_audio.format as wav." ], "material_types": [ "audio" ], "required": [ "model", "input_audio" ] }, "lifecycle": { "assets": [ { "http": "GET", "path": "/v1/media/assets" }, { "http": "GET", "path": "/v1/media/assets/{asset_id}" } ], "content": [ { "http": "GET", "path": "/v1/media/assets/{asset_id}/content", "url_source": "exact returned signed content URL" } ], "dedicated_method_polling_or_cancel": false, "mode": "synchronous_result_then_owned_library", "original_POST_adds_task_envelope": false, "status": [ { "http": "GET", "id_source": "exact task_id from an existing owned media record", "path": "/v1/media/tasks/{task_id}" } ], "unknown": "Retain the actual X-Client-Request-ID and complete response. Read the original owned assets and their exact signed content URLs; never resubmit an uncertain creation." }, "method": { "http": "POST", "path": "/v1/audio/transcriptions" }, "output": { "all_results": "Retain the complete response and headers, every actual output, original returned identifiers and all usage fields. Recover the original operation without submitting creation again.", "family": "text", "fields": [ { "description": null, "name": "text (when present)", "type": null }, { "description": null, "name": "usage (when present)", "type": null } ], "result": "Complete original response body, status and forwarded headers. No universal JSON shape, text field or task envelope is added by this handler." } }, "ctx": null, "declared_pricing_profiles": [ { "dimensions": { "regime": "base" }, "label": "Base token rates", "units": [ { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "input_price", "label": "Input tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "output_price", "label": "Output tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "output_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "out", "unit": "output_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_read_price", "label": "Cache read", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "cached_input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "cached_input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_price", "label": "Cache write", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "generic_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "generic_cache_creation_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_1h_price", "label": "Cache write, 1 hour", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "one_hour_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "one_hour_cache_creation_token" } ] }, { "dimensions": { "quoted_unit": "per input audio second" }, "label": "input audio second", "units": [ { "amount": "0.00015", "amount_unit": "usd_per_exact_native_unit", "billing_input": true, "estimate_only": false, "key": "audio_input_second", "label": "input audio second", "native_amount_usd_per_unit": "0.00015", "native_quantity": 1, "public_unit_label": "per input audio second", "quantity": 1, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "audio_input_second" } ] } ], "examples": [ { "code": "set -eu\n: \"${AILOOT_API_KEY:?Set your Ailoot API key}\"\n: \"${AILOOT_AUDIO_FILE:?Set the path to your actual PCM WAV recording}\"\nexport AILOOT_AUDIO_FILE\npython3 - <<'PY_REQUEST'\nimport base64, io, json, os, wave\nfrom pathlib import Path\naudio = Path(os.environ['AILOOT_AUDIO_FILE']).read_bytes()\nif not (audio[:4] == b'RIFF' and audio[8:12] == b'WAVE'):\n raise SystemExit('Supply your actual PCM WAV audio file')\nwith wave.open(io.BytesIO(audio), 'rb') as recording:\n if recording.getcomptype() != 'NONE' or recording.getnframes() <= 0:\n raise SystemExit('Supply a nonempty PCM WAV recording')\nrequest = {'model': 'openai/gpt-transcribe', 'input_audio': {'data': base64.b64encode(audio).decode('ascii'), 'format': 'wav'}}\nPath('request.json').write_text(json.dumps(request, ensure_ascii=False), encoding='utf-8')\nPY_REQUEST\ncurl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/audio/transcriptions \\\n -H \"Authorization: Bearer $AILOOT_API_KEY\" -H 'Content-Type: application/json' \\\n --data-binary @request.json -D response-headers.txt -o response.bin\n# Retain the complete original response, headers and actual X-Client-Request-ID.\n# Read existing owned records for recovery; do not resubmit an uncertain operation.\n", "label": "curl · actual WAV file and complete original result", "language": "bash" } ], "family": "openai", "fast_multiplier": null, "flex_multiplier": null, "fmt": "transcriptions", "hints": [ "The example reads your actual nonempty PCM WAV file, encodes the complete file bytes as base64 and sends input_audio.format as wav.", "The format value in the example is the documented recipe; no model-specific format list, limit or default is asserted.", "Retain the complete response and headers, every actual output, original returned identifiers and all usage fields. Recover the original operation without submitting creation again." ], "http": "POST", "in": null, "input": [ "audio" ], "label": "Transcription", "max_out": null, "max_reasoning_effort_multiplier": null, "method_aliases": [], "native_endpoint": "audio_transcriptions", "out": null, "output": [ "text" ], "path": "/v1/audio/transcriptions", "price_conditions": [ { "label": "Fast mode", "value": "Not specified", "when": "When fast mode is selected" }, { "label": "Flex mode", "value": "Not specified", "when": "When flex mode is selected" }, { "label": "Maximum reasoning effort", "value": "Not specified", "when": "When maximum reasoning effort uses the default adjustment" } ], "pricing": { "billing_mode": "per_request", "image_input_price": null, "image_output_price": null, "per_request_price": null, "unit_prices": { "audio_input_second": 0.00015 } }, "pricing_profiles": [ { "dimensions": { "regime": "base" }, "label": "Base token rates", "units": [ { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "input_price", "label": "Input tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "output_price", "label": "Output tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "output_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "out", "unit": "output_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_read_price", "label": "Cache read", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "cached_input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "cached_input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_price", "label": "Cache write", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "generic_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "generic_cache_creation_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_1h_price", "label": "Cache write, 1 hour", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "one_hour_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "one_hour_cache_creation_token" } ] }, { "dimensions": { "quoted_unit": "per input audio second" }, "label": "input audio second", "units": [ { "amount": "0.00015", "amount_unit": "usd_per_exact_native_unit", "billing_input": true, "estimate_only": false, "key": "audio_input_second", "label": "input audio second", "native_amount_usd_per_unit": "0.00015", "native_quantity": 1, "public_unit_label": "per input audio second", "quantity": 1, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "audio_input_second" } ] } ], "request_model_id": "openai/gpt-transcribe", "route_label": "Ultra", "status_reads": [ "GET /v1/media/tasks/{task_id}" ], "tiers": [], "units": [ { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "input_price", "label": "Input tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "output_price", "label": "Output tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "output_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "out", "unit": "output_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_read_price", "label": "Cache read", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "cached_input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "cached_input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_price", "label": "Cache write", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "generic_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "generic_cache_creation_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_1h_price", "label": "Cache write, 1 hour", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "one_hour_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "one_hour_cache_creation_token" }, { "amount": "0.00015", "amount_unit": "usd_per_exact_native_unit", "billing_input": true, "estimate_only": false, "key": "audio_input_second", "label": "input audio second", "native_amount_usd_per_unit": "0.00015", "native_quantity": 1, "public_unit_label": "per input audio second", "quantity": 1, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "audio_input_second" } ] } } ], "model_id": "openai/gpt-transcribe", "nativeProvenance": { "input_facts_canonical_sha256": "cfde9e13070d8e615f91456c1b7ef6672cc511be8bb80be8386ce0d253468b32", "input_html_sha256": "b04fd4e8b977b4073c05c30f342d1a8912a8bbdd180fb527d5dcc793207bae6d", "private_audit_sha256": "5ac456b15a9d4cee7020a3b88f7934944c22b965ee99b3c181fa09917ceba28e", "role": "historical-normalized-input-audit-hashes", "schema": "ailoot-native-model-provenance-v2" }, "page": { "canonical": "https://ailoot.ai/models/en/openai-gpt-transcribe/", "description": "GPT Transcribe turns recorded speech into text in its original language, with contextual prompts, literal keyword hints and multiple language hints. OpenAI documents completed-file transcription, streamed file results and transcription after committed Realtime audio turns.", "locale": "en", "md_url": "https://ailoot.ai/models/en/openai-gpt-transcribe/index.md", "title": "OpenAI: GPT Transcribe — Ailoot", "url": "https://ailoot.ai/models/en/openai-gpt-transcribe/" }, "rawModel": { "about": "GPT Transcribe is OpenAI’s speech-to-text model for completed audio files, streamed file transcripts and committed audio turns in Realtime sessions. It accepts audio with optional text context and returns recognized text. The producer positions it for accurate transcription of recorded speech in its original language, including domain vocabulary, multilingual recordings and code-switching. Its useful output is a transcript that stays connected to the recording, with detected-language metadata when the model can make a reliable prediction.\n\n### Preparing a completed recording\n\nOpenAI’s file-transcription workflow accepts uploads up to 25 MB. The current endpoint reference lists FLAC, MP3, MP4, MPEG, MPGA, M4A, OGG, WAV and WebM input formats and recommends enough file metadata to identify the format, including an appropriate filename extension and content type. For a larger recording, the guide recommends compression or splitting it into files of 25 MB or less. Avoid cutting through a sentence, because removing surrounding speech can make recognition less accurate. Keep the pieces in order and retain their positions in the original recording.\n\nFor a meeting, interview or recorded presentation, supply the actual audio and identify the topic and any terminology likely to be spoken. A useful result is the recognized wording rather than an inferred summary of the discussion. Preserve the recording’s identifier with the returned text so that a person reviewing a name, number or uncertain phrase can return to the right source. A transcript can then become material for a separate summary, translation or editorial workflow.\n\n### Context, keyword hints and expected languages\n\nThe producer separates three kinds of guidance. A prompt gives free-form context about the recording’s topic or setting. Keywords give literal words or phrases expected in the speech. A languages list identifies the expected spoken languages. For a bilingual product discussion, describe the setting in the prompt, list the relevant product names as keywords, and provide the expected languages. These are recognition hints; they are not instructions to insert vocabulary that nobody said.\n\nGPT Transcribe uses the plural languages field instead of the older singular language field; OpenAI says not to send both. Its guide accepts supported ISO 639-1 codes, selected ISO 639-3 codes and regional Chinese locale codes, with unsupported or malformed values rejected. Each keyword must stay on one line and must not contain an angle bracket, carriage return or line feed. An invalid keyword or an overlong prompt can reject the complete request. Keep context relevant and check whether hinted names actually occur in the audio rather than treating the hint list as the transcript.\n\n### JSON output and detected-language predictions\n\nThe documented file result is JSON containing the transcript and detected languages. Those language predictions are the model’s output, not a copy of the supplied language hints. An empty language list is a valid result when the model cannot make a reliable prediction; it does not establish that the recording was silent or that transcription failed. Keep the recognized text and language metadata separate so downstream processing can use each according to its meaning.\n\n### Streaming a file transcript\n\nA completed file can produce transcript text incrementally while it is processed. OpenAI documents transcript.text.delta events for partial text and transcript.text.done for the complete file transcript; the final event also includes detected languages. This streams the output of an already supplied recording. It is useful for displaying progress on a long upload, while the final event establishes the text to retain as the completed result.\n\n### Committed turns in Realtime sessions\n\nGPT Transcribe also has a specialized Realtime workflow over WebSocket. Audio is appended to the input buffer, then the application explicitly commits an audio turn. Transcription begins for that committed chunk and may emit deltas before its completion event, including detected-language information. The model automatically uses earlier transcribed turns as context in this producer session flow. This is distinct from continuously emitting live-caption text as microphone audio arrives; OpenAI names GPT Live Transcribe for that continuous workflow.\n\nFor several turns, retain the input-item identifier with the corresponding transcript. OpenAI warns that completion events from different turns need not arrive in speech order, so associate each final result with its committed input rather than arranging a conversation solely by completion time. The useful artifact is an ordered transcript assembled from the correct final items, with interim text treated as progress rather than a separate completed utterance.\n\n### Specialized outputs and substantive limits\n\nThis profile does not supply a universal subtitle, timestamp, translation or speaker-label contract. OpenAI’s migration guide distinguishes those specialized capabilities: native subtitle formats and word timing require a compatible model and format, while speaker labels use GPT-4o Transcribe Diarize. GPT Transcribe’s file result is JSON rather than a drop-in SRT, VTT or verbose-timestamp response. Its exact model card does not publish a numeric context or maximum-output allowance, so a chat-model token window should not be assumed for audio transcription.\n\nReview difficult vocabulary, numbers, accents, background noise and changes of language against the recording. Check that relevant hints improved recognition without introducing unspoken terms and that the final transcript includes the end of the audio. For a multilingual result, preserve the spoken wording before preparing any translated version. These checks make the transcript useful as source material while keeping context hints, detected languages and the recognized speech clearly distinguished.", "alias_of": null, "aliases": [], "author": "OpenAI", "capabilities": { "context_length": 0, "manufacturer_input": [ "audio" ], "manufacturer_output": [ "transcription" ] }, "card_routes": [ { "label": "Ultra", "rates": [ { "amount": "0.00015", "amount_unit": "usd_per_exact_native_unit", "billing_input": true, "estimate_only": false, "key": "audio_input_second", "label": "input audio second", "native_amount_usd_per_unit": "0.00015", "native_quantity": 1, "public_unit_label": "per input audio second", "quantity": 1, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "audio_input_second" } ], "summary": "Full configuration pricing in details." } ], "ctx": null, "formats": [ "transcriptions" ], "id": "openai/gpt-transcribe", "input": [ "audio" ], "kind": "audio", "max_out": null, "methods": [ { "available": null, "cache_read": null, "cache_write": null, "cache_write_1h": null, "contract": { "data": { "all_results": "Retain the complete response and headers, every actual output, original returned identifiers and all usage fields. Recover the original operation without submitting creation again.", "errors": [ { "code": null, "condition": "The JSON body or explicit canonical model ID is invalid.", "http_status": 400, "recovery": "Retain the response and correct the indicated request field.", "type": "invalid_request_error" }, { "code": null, "condition": "The Ailoot API key is invalid.", "http_status": 401, "recovery": "Use your authorized Ailoot API key.", "type": "authentication_error" }, { "code": null, "condition": "The key group does not admit this method.", "http_status": 403, "recovery": "Retain the returned error details.", "type": "permission_error" }, { "code": null, "condition": "The selected method has no admitted route.", "http_status": 404, "recovery": "Retain the returned error details.", "type": "not_found_error" }, { "code": null, "condition": "The accepted tariff cannot obtain its required input quantity or bound.", "http_status": 422, "recovery": "Retain the response; use the input representation required by the accepted tariff.", "type": "price_unavailable" }, { "code": null, "condition": "Required storage, route, method price or reservation is unavailable.", "http_status": 503, "recovery": "Keep the complete original response. An uncertain reservation is not permission to resubmit.", "type": null }, { "code": null, "condition": "The original result, usage or settlement requires checking.", "http_status": 502, "operation_id_source": "actual X-Client-Request-ID", "recovery": "Retain the actual request ID and original response. Read the existing owned record; do not resubmit creation.", "response_headers": { "x-should-retry": "false" }, "type": "server_error" } ], "lifecycle": { "assets": [ { "http": "GET", "path": "/v1/media/assets" }, { "http": "GET", "path": "/v1/media/assets/{asset_id}" } ], "content": [ { "http": "GET", "path": "/v1/media/assets/{asset_id}/content", "url_source": "exact returned signed content URL" } ], "dedicated_method_polling_or_cancel": false, "mode": "synchronous_result_then_owned_library", "original_POST_adds_task_envelope": false, "status": [ { "http": "GET", "id_source": "exact task_id from an existing owned media record", "path": "/v1/media/tasks/{task_id}" } ], "unknown": "Retain the actual X-Client-Request-ID and complete response. Read the original owned assets and their exact signed content URLs; never resubmit an uncertain creation." }, "outputs": { "all_results": "Retain the complete response and headers, every actual output, original returned identifiers and all usage fields. Recover the original operation without submitting creation again.", "family": "text", "fields": [ { "description": null, "name": "text (when present)", "type": null }, { "description": null, "name": "usage (when present)", "type": null } ], "result": "Complete original response body, status and forwarded headers. No universal JSON shape, text field or task envelope is added by this handler." }, "parameters": [ { "default_known": false, "enum": [ "openai/gpt-transcribe" ], "enum_known": true, "name": "model", "required": true, "type": "string" }, { "default_known": false, "enum_known": false, "name": "input_audio", "required": true, "type": "object" } ], "request_schema": { "properties": { "input_audio": { "properties": { "data": { "contentEncoding": "base64", "type": "string" }, "format": { "type": "string" } }, "required": [ "data", "format" ], "type": "object" }, "model": { "enum": [ "openai/gpt-transcribe" ], "type": "string" } }, "required": [ "model", "input_audio" ], "type": "object" }, "schema_template": { "actual_file_preparation_required": true, "body": { "input_audio": { "data": "${BASE64_ACTUAL_WAV_BYTES}", "format": "wav" }, "model": "openai/gpt-transcribe" }, "content_type": "application/json" } }, "input": { "content_type": "application/json", "material_conditions": [ "The example reads your actual nonempty PCM WAV file, encodes the complete file bytes as base64 and sends input_audio.format as wav." ], "material_types": [ "audio" ], "required": [ "model", "input_audio" ] }, "lifecycle": { "assets": [ { "http": "GET", "path": "/v1/media/assets" }, { "http": "GET", "path": "/v1/media/assets/{asset_id}" } ], "content": [ { "http": "GET", "path": "/v1/media/assets/{asset_id}/content", "url_source": "exact returned signed content URL" } ], "dedicated_method_polling_or_cancel": false, "mode": "synchronous_result_then_owned_library", "original_POST_adds_task_envelope": false, "status": [ { "http": "GET", "id_source": "exact task_id from an existing owned media record", "path": "/v1/media/tasks/{task_id}" } ], "unknown": "Retain the actual X-Client-Request-ID and complete response. Read the original owned assets and their exact signed content URLs; never resubmit an uncertain creation." }, "method": { "http": "POST", "path": "/v1/audio/transcriptions" }, "output": { "all_results": "Retain the complete response and headers, every actual output, original returned identifiers and all usage fields. Recover the original operation without submitting creation again.", "family": "text", "fields": [ { "description": null, "name": "text (when present)", "type": null }, { "description": null, "name": "usage (when present)", "type": null } ], "result": "Complete original response body, status and forwarded headers. No universal JSON shape, text field or task envelope is added by this handler." } }, "ctx": null, "declared_pricing_profiles": [ { "dimensions": { "regime": "base" }, "label": "Base token rates", "units": [ { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "input_price", "label": "Input tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "output_price", "label": "Output tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "output_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "out", "unit": "output_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_read_price", "label": "Cache read", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "cached_input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "cached_input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_price", "label": "Cache write", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "generic_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "generic_cache_creation_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_1h_price", "label": "Cache write, 1 hour", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "one_hour_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "one_hour_cache_creation_token" } ] }, { "dimensions": { "quoted_unit": "per input audio second" }, "label": "input audio second", "units": [ { "amount": "0.00015", "amount_unit": "usd_per_exact_native_unit", "billing_input": true, "estimate_only": false, "key": "audio_input_second", "label": "input audio second", "native_amount_usd_per_unit": "0.00015", "native_quantity": 1, "public_unit_label": "per input audio second", "quantity": 1, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "audio_input_second" } ] } ], "examples": [ { "code": "set -eu\n: \"${AILOOT_API_KEY:?Set your Ailoot API key}\"\n: \"${AILOOT_AUDIO_FILE:?Set the path to your actual PCM WAV recording}\"\nexport AILOOT_AUDIO_FILE\npython3 - <<'PY_REQUEST'\nimport base64, io, json, os, wave\nfrom pathlib import Path\naudio = Path(os.environ['AILOOT_AUDIO_FILE']).read_bytes()\nif not (audio[:4] == b'RIFF' and audio[8:12] == b'WAVE'):\n raise SystemExit('Supply your actual PCM WAV audio file')\nwith wave.open(io.BytesIO(audio), 'rb') as recording:\n if recording.getcomptype() != 'NONE' or recording.getnframes() <= 0:\n raise SystemExit('Supply a nonempty PCM WAV recording')\nrequest = {'model': 'openai/gpt-transcribe', 'input_audio': {'data': base64.b64encode(audio).decode('ascii'), 'format': 'wav'}}\nPath('request.json').write_text(json.dumps(request, ensure_ascii=False), encoding='utf-8')\nPY_REQUEST\ncurl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/audio/transcriptions \\\n -H \"Authorization: Bearer $AILOOT_API_KEY\" -H 'Content-Type: application/json' \\\n --data-binary @request.json -D response-headers.txt -o response.bin\n# Retain the complete original response, headers and actual X-Client-Request-ID.\n# Read existing owned records for recovery; do not resubmit an uncertain operation.\n", "label": "curl · actual WAV file and complete original result", "language": "bash" } ], "family": "openai", "fast_multiplier": null, "flex_multiplier": null, "fmt": "transcriptions", "hints": [ "The example reads your actual nonempty PCM WAV file, encodes the complete file bytes as base64 and sends input_audio.format as wav.", "The format value in the example is the documented recipe; no model-specific format list, limit or default is asserted.", "Retain the complete response and headers, every actual output, original returned identifiers and all usage fields. Recover the original operation without submitting creation again." ], "http": "POST", "in": null, "input": [ "audio" ], "label": "Transcription", "max_out": null, "max_reasoning_effort_multiplier": null, "method_aliases": [], "native_endpoint": "audio_transcriptions", "out": null, "output": [ "text" ], "path": "/v1/audio/transcriptions", "price_conditions": [ { "label": "Fast mode", "value": "Not specified", "when": "When fast mode is selected" }, { "label": "Flex mode", "value": "Not specified", "when": "When flex mode is selected" }, { "label": "Maximum reasoning effort", "value": "Not specified", "when": "When maximum reasoning effort uses the default adjustment" } ], "pricing": { "billing_mode": "per_request", "image_input_price": null, "image_output_price": null, "per_request_price": null, "unit_prices": { "audio_input_second": 0.00015 } }, "pricing_profiles": [ { "dimensions": { "regime": "base" }, "label": "Base token rates", "units": [ { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "input_price", "label": "Input tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "output_price", "label": "Output tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "output_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "out", "unit": "output_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_read_price", "label": "Cache read", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "cached_input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "cached_input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_price", "label": "Cache write", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "generic_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "generic_cache_creation_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_1h_price", "label": "Cache write, 1 hour", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "one_hour_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "one_hour_cache_creation_token" } ] }, { "dimensions": { "quoted_unit": "per input audio second" }, "label": "input audio second", "units": [ { "amount": "0.00015", "amount_unit": "usd_per_exact_native_unit", "billing_input": true, "estimate_only": false, "key": "audio_input_second", "label": "input audio second", "native_amount_usd_per_unit": "0.00015", "native_quantity": 1, "public_unit_label": "per input audio second", "quantity": 1, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "audio_input_second" } ] } ], "request_model_id": "openai/gpt-transcribe", "route_label": "Ultra", "status_reads": [ "GET /v1/media/tasks/{task_id}" ], "tiers": [], "units": [ { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "input_price", "label": "Input tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "output_price", "label": "Output tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "output_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "out", "unit": "output_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_read_price", "label": "Cache read", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "cached_input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "cached_input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_price", "label": "Cache write", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "generic_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "generic_cache_creation_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_1h_price", "label": "Cache write, 1 hour", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "one_hour_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "one_hour_cache_creation_token" }, { "amount": "0.00015", "amount_unit": "usd_per_exact_native_unit", "billing_input": true, "estimate_only": false, "key": "audio_input_second", "label": "input audio second", "native_amount_usd_per_unit": "0.00015", "native_quantity": 1, "public_unit_label": "per input audio second", "quantity": 1, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "audio_input_second" } ] } ], "model_aliases": [ { "canonical_model_id": "openai/gpt-transcribe", "request_model_id": "openai/gpt-transcribe", "route_label": "Ultra" } ], "name": "OpenAI: GPT Transcribe", "price": { "in": null, "in_label": null, "out": null, "out_label": null, "sort_in": null, "sort_out": null, "varies": true }, "short": "GPT Transcribe turns recorded speech into text in its original language, with contextual prompts, literal keyword hints and multiple language hints. OpenAI documents completed-file transcription, streamed file results and transcription after committed Realtime audio turns." }, "renderer_schema": { "method_order": "Original DTO methods order; every exact route/format/path/caller survives.", "prices": "Original public DTO fields and original native display HTML; no flattened or normalized offers." }, "representation": "exact-public-dto-and-native-renderer-html", "representationNotes": { "current_reference": { "about_json_pointer": "/aboutHtml", "methods": [ { "contract_json_pointer": "/methods/0/rawMethod/contract", "html_json_pointer": "/methods/0/html", "method_ordinal": 0 } ], "role": "current-semantic-reference" }, "raw_input": { "examples_role": "historical-input-not-current-ready-examples", "json_pointers": [ "/rawModel", "/methods/0/rawMethod" ], "role": "historical-owner-input" }, "schema": "ailoot-native-model-representations-v1" }, "schema": "ailoot-static-model-facts-v1", "schema_version": 1, "shortText": "GPT Transcribe turns recorded speech into text in its original language, with contextual prompts, literal keyword hints and multiple language hints. OpenAI documents completed-file transcription, streamed file results and transcription after committed Realtime audio turns.", "source": { "audit_sha256": "2e00086bcbfc28e64d0cebe97aa47c17d8d5618963027f8ffb9dc6123d7d25b1", "bridge_map_sha256": "1125275768b9636fc8461bc7da5061239a81ddfe508e2557843ca42c58ab0036", "raw_model_json_sha256": "45bce6e96edea61a1e5e5532b85dfeacd09b009549c51153dadb2c3264df663e", "renderer_sha256": "fed50305bd3bcc4980ac14df23640efb8c1bc1e819f9d53643d6f43e8974d3fe" }, "sourceRevision": "4c964c0a2c4f335035b4aa0a9bf6008f0776847cb5c54a823a55b0ec288d8f99" }