문장 인식

API Description
The One-Sentence Recognition API transcribes short audio files up to 60 seconds in duration.
Supported audio formats: wav, pcm, ogg-opus, mp3, m4a.
Requests must use HTTP POST with Content-Type: application/json; charset=utf-8.
API Requirements
Follow these requirements when integrating the One-Sentence Recognition API.
Item
Description
Supported Languages
Including zh (Chinese), en (English), yue (Cantonese), ar (Arabic), de (German), fr (French), es (Spanish), pt (Portuguese), id (Indonesian), it (Italian), ko (Korean), ru (Russian), th (Thai), vi (Vietnamese), ja (Japanese), tr (Turkish), hi (Hindi), ms (Malay), nl (Dutch), sv (Swedish), da (Danish), fi (Finnish), pl (Polish), cs (Czech), fil (Filipino), fa (Persian), el (Greek), ro (Romanian), hu (Hungarian), and mk (Macedonian).
Supported Industries
General purpose.
Audio Properties
Sample rate: 16000Hz (pcm format supports 8000Hz via the InputSampleRate parameter; see Input Parameters section);
Sample precision: 16 bits; 
Channel: mono.
Audio Formats
wav, pcm, ogg-opus, mp3, m4a.
Request Protocol
HTTPS protocol.
Request URL
https://asr-intl.cloud-rtc.com/v1/SentenceRecognition?{request parameters}.
Response Format
JSON.
Data Submission
Supports two methods: uploading a local audio file or providing an audio URL. Audio duration must not exceed 60 seconds, and file size must not exceed 3MB.
Concurrency Limit
The default concurrency limit per account is 30 requests per second.
Note：
Under the same SDKAppID, all real-time speech recognition scenarios—including direct STT access, AI Conversation, and AI Transcription/Translation—share a single concurrency pool. This means the total number of concurrent streams across all types cannot exceed the account's upper limit. One-Sentence Recognition and Audio File Recognition are calculated independently and do not consume the real-time recognition concurrency quota. If you require higher concurrency,  please Contact Us.
Workflow
﻿
API Call Flow
Request Format
To call the API, send an HTTP POST request. The production URL format is:
https://asr-intl.cloud-rtc.com/v1/SentenceRecognition?{request parameters}
Add the following fields to the request header:
key
Required
Value Description
X-TRTC-SdkAppId
Yes
TRTC Application ID. See Application Basic Information for instructions. Example: 1400188366.
X-TRTC-UserSig
Yes
TRTC Signature. See User Authentication for calculation and usage. The UserID here matches the RequestId parameter in the URL.
X-Business-Name
No
Leave empty unless instructed by the platform after internal business communication. Do not fill arbitrarily.
Request Parameters
Request parameters format: AppId=value1&RequestId=value2
Parameter details:
Parameter Name
Required
Type
Description
AppId
Yes
String
Tencent Cloud account AppId. Obtain from API SecretKey Management.
RequestId
Yes
String
Globally unique request ID (UUID). Used to generate the UserSig for this request.
Input Parameters
The request body must be in JSON format.
Parameter details:
Parameter Name
Required
Type
Description
EngServiceType
Yes
String
Engine model type. Supported value:
bigmodel: Large model edition.
SourceType
Yes
Integer
Audio data source: 
0: Audio URL;
1: Audio data (request body).
VoiceFormat
Yes
String
Audio format for recognition. Supported formats: wav, pcm, ogg-opus, mp3, m4a.
Url
No
String
Audio URL. Must be publicly accessible. Required when SourceType is 0; not needed when SourceType is 1. Audio duration must not exceed 60 seconds, file size must not exceed 3MB.
Data
No
String
Audio data. Required for SourceType 1 (local upload); optional for SourceType 0 (audio URL). Must be Base64 encoded. (If using Python, ensure the file is read as a string, not bytes. If read as bytes, use decode(). Encoded data must not contain line breaks.) Audio duration must not exceed 60 seconds, and file size must not exceed 3MB (after Base64 encoding).
DataLen
No
Integer
Length of the audio data in bytes. Required for SourceType 1 (local upload); optional for SourceType 0 (audio URL). This is the length before Base64 encoding.
WordInfo
No
Integer
Controls word-level timestamp display: 
0: Do not display;
1: Display, exclude punctuation timestamps;
2: Display, include punctuation timestamps. Default: 0.
FilterDirty
No
Integer
Profanity filter (Mandarin Chinese): 
0: Do not filter;
1: Filter;
2: Replace profanities with *. Default: 0.
FilterModal
No
Integer
Modal word filter (Mandarin Chinese): 
0: Do not filter;
1: Partial filter;
2: Strict filter. Default: 0.
FilterPunc
No
Integer
Punctuation filter (Mandarin Chinese): 
0: Do not filter;
1: Filter sentence-ending punctuation;
2: Filter all punctuation. Default: 0.
ConvertNumMode
No
Integer
Smart conversion to Arabic numerals: 
0: Output Chinese numerals directly;
1: Convert to Arabic numerals based on context. Default: 1.
HotwordList
No
String
Temporary hotword list to improve recognition accuracy.
Single hotword format: "hotword|weight". Each hotword must not exceed 30 characters (or 10 Chinese characters). Supported weights: 1-11 or 100. For example: "Tencent|5" or "ASR|11".
Multiple hotwords must be separated by commas. A maximum of 128 hotwords is supported. For example: "hotword|10,Tencent|5,ASR|11".
Note：
A weight of 11 upgrades the hotword to a super hotword. Use this weight only for important terms that must take effect; configuring too many super hotwords may reduce overall recognition accuracy.
A weight of 100 enables enhanced homophone replacement. For example, if "site|100" is configured, recognition results such as "sight" or "cite", which share the same pronunciation, are forcibly replaced with "site". Use this setting only for important terms that must take effect; configuring too many such hotwords may reduce overall recognition accuracy.
Hotwords cannot contain spaces. For example, use "OpenAIAPI" rather than "OpenAI API".
﻿
InputSampleRate
No
Integer
Supports upsampling 8k pcm audio to 16k for recognition when the engine sample rate does not match, improving accuracy. Only supports: 8000. For example, if 8000 is passed, pcm audio sample rate is 8k; when using the 16k_zh engine, 8k pcm audio can be recognized normally. 
Note：
This parameter applies only to pcm audio. If not provided, default behavior is used (engine sample rate matches pcm audio sample rate).
Language
No
String
When EngServiceType is set to bigmodel, you can specify the audio language. Supported languages: zh (Chinese), en (English), yue (Cantonese), ar (Arabic), de (German), fr (French), es (Spanish), pt (Portuguese), id (Indonesian), it (Italian), ko (Korean), ru (Russian), th (Thai), vi (Vietnamese), ja (Japanese), tr (Turkish), hi (Hindi), ms (Malay), nl (Dutch), sv (Swedish), da (Danish), fi (Finnish), pl (Polish), cs (Czech), fil (Filipino), fa (Persian), el (Greek), ro (Romanian), hu (Hungarian), mk (Macedonian). Example value: zh.
Note:
If this field is left empty, the model will automatically detect the language.
Response Format
Responses are returned in JSON format. All fields are contained within the top-level Response object.
Parameter details:
Parameter Name
Type
Description
Response
Object
Top-level response object containing the following fields.
Response.Result
String
Recognition result.
Response.AudioDuration
Integer
Audio duration of the request, in milliseconds.
Response.WordSize
Integer
Length of the word timestamp list. 
Note:
This field may return null if no valid value is available.
Response.WordList
Array of SentenceWord
Word timestamp list. 
Note:
This field may return null if no valid value is available.
Response.RequestId
String
RequestId for this request. Provide this RequestId when troubleshooting issues.
SentenceWord definition:
Name
Type
Description
Word
String
Recognized word.
StartTime
Integer
Start time of the word in the audio.
EndTime
Integer
End time of the word in the audio.
Example
Here is an example of a One-Sentence Recognition request with SourceType=1:
Request URL:
https://asr.cloud-rtc.com/v1/SentenceRecognition?AppId=12345&RequestId=2f5e8263-a23b-4e65-be22-6b7c19bd4064
Input parameters:
{
    "EngSerViceType": "bigmodel",
    "SourceType": 1,
    "VoiceFormat": "wav",
    "Data": "xxxxxxxx",
    "DataLen": 8
}
Response:
{
    "Response": {
        "Result": "Within what we can accept.",
        "RequestId": "2f5e8263-a23b-4e65-be22-6b7c19bd4064",
        "AudioDuration": 1840
    }
}
Developer Resources
SDK
Tencent-RTC SDK for go: Github﻿
Tencent-RTC SDK for python: Github﻿
Tencent-RTC SDK for nodejs: Github﻿
Tencent-RTC SDK for java: Github﻿
Tencent-RTC SDK for rust: Github﻿
Tencent-RTC SDK for cpp: Github﻿
SDK Sample Code
﻿Golang﻿
﻿Python﻿
﻿Node.js﻿
﻿Java﻿
﻿Rust﻿
﻿CPP﻿
Error Codes
Code
Description
400
Audio upload failed.
413
Audio file exceeds size or duration limit.
400
Audio file is empty.
400
Request body format error.
4002
Authentication failed.
4003
AppID service not enabled. Please enable the service in the console.
4004
Resource Package exhausted. Please enable Pay-as-you-go or purchase a Resource Package.
4005
Account suspended due to overdue payment. Please recharge promptly.
4006
Account concurrency limit exceeded.
4007
Audio decoding failed. Please verify that the uploaded audio format matches the request parameters.
5000
Recognition failed due to high machine load or network jitter. Please retry. 
Note:
This issue is typically sporadic and can be ignored if infrequent. Simply retry recognition.
5001
Recognition failed due to high machine load or network jitter. Please retry. 
Note:
This issue is typically sporadic and can be ignored if infrequent. Simply retry recognition.
5002
Recognition failed due to high machine load or network jitter. Please retry. 
Note:
This issue is typically sporadic and can be ignored if infrequent. Simply retry recognition.

Item	Description
Supported Languages	Including zh (Chinese), en (English), yue (Cantonese), ar (Arabic), de (German), fr (French), es (Spanish), pt (Portuguese), id (Indonesian), it (Italian), ko (Korean), ru (Russian), th (Thai), vi (Vietnamese), ja (Japanese), tr (Turkish), hi (Hindi), ms (Malay), nl (Dutch), sv (Swedish), da (Danish), fi (Finnish), pl (Polish), cs (Czech), fil (Filipino), fa (Persian), el (Greek), ro (Romanian), hu (Hungarian), and mk (Macedonian).
Supported Industries	General purpose.
Audio Properties	Sample rate: 16000Hz (pcm format supports 8000Hz via the InputSampleRate parameter; see Input Parameters section); Sample precision: 16 bits; Channel: mono.
Audio Formats	wav, pcm, ogg-opus, mp3, m4a.
Request Protocol	HTTPS protocol.
Request URL	`https://asr-intl.cloud-rtc.com/v1/SentenceRecognition?{request parameters}`.
Response Format	JSON.
Data Submission	Supports two methods: uploading a local audio file or providing an audio URL. Audio duration must not exceed 60 seconds, and file size must not exceed 3MB.
Concurrency Limit	The default concurrency limit per account is 30 requests per second.

key	Required	Value Description
X-TRTC-SdkAppId	Yes	TRTC Application ID. See Application Basic Information for instructions. Example: 1400188366.
X-TRTC-UserSig	Yes	TRTC Signature. See User Authentication for calculation and usage. The UserID here matches the RequestId parameter in the URL.
X-Business-Name	No	Leave empty unless instructed by the platform after internal business communication. Do not fill arbitrarily.

Parameter Name	Required	Type	Description
AppId	Yes	String	Tencent Cloud account AppId. Obtain from API SecretKey Management.
RequestId	Yes	String	Globally unique request ID (UUID). Used to generate the UserSig for this request.

Name	Type	Description
Word	String	Recognized word.
StartTime	Integer	Start time of the word in the audio.
EndTime	Integer	End time of the word in the audio.

Code	Description
400	Audio upload failed.
413	Audio file exceeds size or duration limit.
400	Audio file is empty.
400	Request body format error.
4002	Authentication failed.
4003	AppID service not enabled. Please enable the service in the console.
4004	Resource Package exhausted. Please enable Pay-as-you-go or purchase a Resource Package.
4005	Account suspended due to overdue payment. Please recharge promptly.
4006	Account concurrency limit exceeded.
4007	Audio decoding failed. Please verify that the uploaded audio format matches the request parameters.
5000	Recognition failed due to high machine load or network jitter. Please retry. Note: This issue is typically sporadic and can be ignored if infrequent. Simply retry recognition.
5001	Recognition failed due to high machine load or network jitter. Please retry. Note: This issue is typically sporadic and can be ignored if infrequent. Simply retry recognition.
5002	Recognition failed due to high machine load or network jitter. Please retry. Note: This issue is typically sporadic and can be ignored if infrequent. Simply retry recognition.