Fun-ASR audio file transcription iOS SDK

Calling steps

Connection and control parameters

You can configure these by passing a JSON string to the <a baseurl="t3182910_v1_0_0.xdita" data-node="6224415" data-root="85177" data-tag="xref" href="#05eab5125e2pm" id="ea3603d9fd8f3">nui_initialize</a> interface's parameters parameter.

• Parameter example: The following shows a sample JSON string. Not all parameters are listed. Add any missing parameters as needed:

  {
      "url": "wss://dashscope.aliyuncs.com/api-ws/v1/inference",
      "apikey": "st-****",
      "device_id": "my_device_id",
      "service_mode": "1"
  }

• Parameter description

  Parameter	Type	Required	Description
  url	String	Yes	Service endpoint. Fixed as wss://dashscope.aliyuncs.com/api-ws/v1/inference.
  apikey	String	Yes	API key.
  service_mode	String	Yes	Operation mode. For audio file transcription, this must be "1".
  device_id	String	Yes	A unique string that identifies the end user. Set this to the user ID within your app or a device-specific identifier generated by the client. This ID helps with log tracing and troubleshooting.
  debug_path	String	No	Path where log files are stored. This parameter takes effect only if you set save_log to YES when calling nui_initialize. In that case, you must specify a log file path. Otherwise, an error occurs. The system keeps at most two local log files.
  max_log_file_size	int	No	Maximum size (in bytes) for a log file. This parameter takes effect only if you set save_log to YES when calling nui_initialize. Default value: 104857600 (100 × 1024 × 1024 bytes, or 100 MiB).
  log_track_level	int	No	Controls the filter level for logs sent externally through the log callback (onFileTransLogTrackCallback). Default value: 2. Valid values: 0: LOG_LEVEL_VERBOSE 1: LOG_LEVEL_DEBUG 2: LOG_LEVEL_INFO 3: LOG_LEVEL_WARNING 4: LOG_LEVEL_ERROR 5: LOG_LEVEL_NONE (disables this feature) Note: log_track_level and level (set via the nui_initialize interface) jointly determine which logs are finally sent in the callback. A log message appears in the callback only if its level is greater than or equal to both log_track_level and level. For example, if log_track_level is set to 2 (INFO) and level is set to 3 (WARNING), only WARNING-level logs (value ≥ 3) and higher appear in the callback.

Speech recognition effect parameters

You can use the <a baseurl="t3182910_v1_0_0.xdita" data-node="6224415" data-root="85177" data-tag="xref" href="#763672f3f8dgw" id="0ab38e798dtrt">nui_set_params</a> interface to configure the nl_config parameter, or use the <a baseurl="t3182910_v1_0_0.xdita" data-node="6224415" data-root="85177" data-tag="xref" href="#8fe6ea298apzu" id="f5ea550ae12nx">nui_file_trans_start</a> interface to configure all speech recognition effect parameters.

• Parameter example: The following shows a sample JSON string. Not all parameters are listed. Add any missing parameters as needed:

  {
      "file_urls": [
          "https://dashscope.oss-cn-beijing.aliyuncs.com/samples/audio/paraformer/hello_world_female2.wav"
      ],
      "async_request": false,
      "nls_config": {
          "model":"fun-asr",
          "diarization_enabled": false,
          "parameters": {
              "speech_noise_threshold": 0.0
          }
      }
  }

• Parameter description

  Parameter	Type	Required	Description
  file_urls	array[string]	Yes	List of URLs for audio or video files to transcribe. Supports HTTP and HTTPS protocols. A single request supports only 1 URL. If your audio files are stored in OSS, the SDK does not support temporary URLs that start with the oss:// prefix. Audio formats: aac, amr, avi, flac, flv, m4a, mkv, mov, mp3, mp4, mpeg, ogg, opus, wav, webm, wma, wmv Important Because audio and video formats and their variants are numerous, it is impossible to test them all. The API cannot guarantee correct recognition for every format. Test your files to confirm they produce normal speech recognition results. Audio sampling rate: Any Audio file size and duration: Files must not exceed 2 GB and must be under 12 hours long. If your files exceed these limits, preprocess them to reduce their size. For best practices on file preprocessing, see Preprocess video files to improve transcription efficiency (for audio file transcription scenarios).
  async_request	boolean	No	Whether the speech recognition request is asynchronous. Default value: false. Valid values: true: Asynchronous request false: Synchronous request
  apikey	string	No	If the apikey in Connection and control parameters uses a temporary API key, update it here to prevent expiration.
  nls_config	object	Yes	Core configuration object for speech recognition. Contains key parameters such as model selection and recognition effect controls.
  nls_config.model	string	Yes	Speech recognition model.
  nls_config.special_word_filter	object	No	Specifies sensitive words to handle during speech recognition and supports different handling methods for each word. If you omit this parameter, the system uses built-in sensitive word filtering logic. Words matching the Alibaba Cloud Model Studio sensitive word list are replaced with asterisks (*) of equal length. If you provide this parameter, you can apply the following sensitive word handling strategies: Replace with *: Replaces matched sensitive words with the same number of * characters. Filter out completely: Matching sensitive words are removed entirely from the recognition result. The value must be a JSON object with the following structure: { "filter_with_signed": { "word_list": ["test"] }, "filter_with_empty": { "word_list": ["start", "occur"] }, "system_reserved_filter": true } JSON field descriptions: filter_with_signed Type: Object. Required: No. Description: The list of sensitive words to be replaced with *. Matched words in the recognition result are replaced by * characters of the same length. Example: With the JSON above, the phrase “help me test this code” becomes “help me ** this code”. Inner fields: word_list: Array of strings listing sensitive words to replace. filter_with_empty Type: Object. Required: No. Description: Configures a list of sensitive words to remove (filter out) from the recognition result. Matching words are deleted entirely. Example: With the JSON above, the phrase “is the match about to start?” becomes “is the match about to?”. Inner fields: word_list: Array of strings listing sensitive words to remove completely. system_reserved_filter Type: Boolean. Required: No. Default value: true. Description: Whether to enable the system’s preset sensitive word rules. If set to true, the system also applies its built-in sensitive word filtering logic. Words matching the Alibaba Cloud Model Studio sensitive word list are replaced with asterisks (*) of equal length.
  nls_config.channel_id	array[integer]	No	Indexes of sound channels to recognize in a multi-channel audio file. The index starts from 0. For example, [0] recognizes the first channel, and [0, 1] recognizes the first and second channels. If omitted, the first channel is processed by default. Important Each specified sound channel is billed separately. For example, a request for [0, 1] for a single file incurs two separate charges. Default value: [0]
  nls_config.diarization_enabled	boolean	No	Automatic speaker diarization is disabled by default. This feature applies to single-channel audio only (not supported for multi-channel audio). When enabled, recognition results include the speaker_id field to distinguish speakers. Note If speaker diarization is enabled, keep the audio duration under 2 hours. Audio exceeding 2 hours may cause recognition failures or timeouts. For an example of speaker_id, see Recognition result description.
  nls_config.speaker_count	integer	No	Reference value for the number of speakers. To use this feature, set diarization_enabled to true. By default, the system automatically determines the number of speakers. If you set this parameter, it only guides the algorithm to aim for the specified number—it does not guarantee that exact number. Valid range: [2, 100]. Because this feature distinguishes multiple speakers, the minimum value is 2.
  nls_config.vocabulary_id	string	No	ID of a hotword vocabulary list to improve recognition accuracy for specific terms. This parameter applies to v2 and later models. For instructions on using hotwords, see Customize hotwords.
  nls_config.language_hints	array[string]	No	The language code for recognition. If the source language is unknown, leave it unset and the model detects the language automatically. The system reads only the first value in the array. Any extra values are ignored. Click to view the supported language codes fun-asr, fun-asr-2025-11-07, fun-asr-mtl, fun-asr-mtl-2025-08-25: zh: Chinese en: English ja: Japanese ko: Korean vi: Vietnamese th: Thai id: Indonesian ms: Malay tl: Filipino hi: Hindi ar: Arabic fr: French de: German es: Spanish pt: Portuguese ru: Russian it: Italian nl: Dutch sv: Swedish da: Danish fi: Finnish no: Norwegian el: Greek pl: Polish cs: Czech hu: Hungarian ro: Romanian bg: Bulgarian hr: Croatian sk: Slovak fun-asr-2025-08-25: zh: Chinese en: English
  nls_config.parameters	object	No	Configures additional parameters as a JSON object.

NeoNui

NeoNuiSdkDelegate: Callback listeners

-(void)onFileTransLogTrackCallback:(NuiSdkLogLevel)level
                        logMessage:(const char *)log;

NuiCallbackEvent: Event types