hi_mpi_ai_enable_resample

The Atlas 200/300/500 Inference Product does not support this API.

The Atlas Training Series Product does not support this API.

Enables AI resampling.

After enabling an AI channel by calling hi_mpi_ai_enable_chn and before binding the VENC channel to the AI channel by calling hi_mpi_sys_bind, call this API to enable resampling.
After an AI channel is disabled by calling hi_mpi_ai_disable_chn, if you want to enable the AI channel again and use the resampling function, you need to call this API to enable resampling again.
Due to the limitation of the Nyquist sampling theorem, when the audio sampling rate is 8 kHz, the supported audio frequency is less than 4 kHz.
The AI sampling rate is fixed at 48 kHz. The G.711a and G.711u protocols only support the sampling rate of 8 kHz, and when either protocol is used, resampling needs to be enabled. However, the AAC protocol supports the sampling rate of 48 kHz, and when it is used, AI resampling is not required.

hi_s32 hi_mpi_ai_enable_resample(hi_audio_dev ai_dev, hi_ai_chn ai_chn, hi_audio_sample_rate out_sample_rate);

Parameter	Input/Output	Description
ai_dev	Input	Audio device ID. The value range is [0, 2]. In the network video recorder (NVR) scenario, only dev2 is involved.
ai_chn	Input	AI channel ID. Only the mono audio channel (audio-left channel) is supported. The value of ai_chn is fixed at 0.
out_sample_rate	Input	Output sampling rate for audio resampling. Only 8 kHz is supported.

Parameter

Input/Output

Description

ai_dev

Input

Audio device ID. The value range is [0, 2].

In the network video recorder (NVR) scenario, only dev2 is involved.

ai_chn

Input

AI channel ID.

Only the mono audio channel (audio-left channel) is supported. The value of ai_chn is fixed at 0.

out_sample_rate

Input

Output sampling rate for audio resampling. Only 8 kHz is supported.

Parent topic: AI/AO