@Julien7720 wrote:
Hello, we are facing a call transcription problem, we have an automatic speaker mapping problem, we have a speaker 1 speaker 2, we have to manually choose which one
Before the callers are automatically assigned with the call participants (User and contact).We are using api to upload record file
You may need to adjust the settings or parameters of your transcription API to ensure that speakers are accurately mapped to the correct callers. This could involve configuring speaker recognition features or refining the algorithms used for speaker identification.
Additionally, you might consider preprocessing the audio files before uploading them to the API to improve speaker recognition accuracy. This could involve segmenting the audio based on different speakers or enhancing the audio quality to facilitate better identification.
