Forum Discussion
Whisper-1 Model Transcribes English Audio Incorrectly
Hi everyone,
I'm currently working with the gpt-4o-realtime-preview model from Azure OpenAI and using the whisper-1 model for audio-to-text transcription. However, I'm encountering a recurring issue where the transcription frequently fails to detect the correct language.
Even though I provide clear English audio, the output is often transcribed in other languages such as Hindi, Urdu, or Chinese. This inconsistency is affecting the reliability of the transcription process.
Here’s a snippet of the code I’m using:
ConversationSessionOptions sessionOptions = new()
{
Voice = ConversationVoice.Alloy,
InputAudioFormat = ConversationAudioFormat.Pcm16,
OutputAudioFormat = ConversationAudioFormat.Pcm16,
Instructions = instructions,
InputTranscriptionOptions = new()
{
Model = "whisper-1",
},
};
Is there a way to explicitly specify or prompt the whisper-1 model to prioritize or lock in English as the transcription language? Any guidance on how to improve language detection accuracy would be greatly appreciated.
Thanks in advance!
Tag
Like
1 Reply
- hazemBrass ContributorYou can force Whisper-1 to lock onto English by explicitly passing a "language": "en" parameter in your transcription request. Since the current Azure.AI.OpenAI SDK (v2.2.0-beta.4) doesn’t yet expose a Language property, you’ll need to call the REST endpoint directly. POST https://{your-resource-name}.openai.azure.com/openai/deployments/whisper-1/audio/transcriptions?api-version=2024-02-15-preview Content-Type: application/json Authorization: Bearer {your-key} { "file": "<base64-or-stream-of-your-audio>", "model": "whisper-1", "language": "en" }
 This tells Whisper to skip auto-detection and transcribe everything as English
 But, If you’d rather stay in the SDK, keep an eye out for the next preview of Azure.AI.OpenAI (v2.3.0+), where ConversationInputTranscriptionOptionsInputTranscriptionOptions = new() { Model = "whisper-1", Language = "en" };Once that lands, you can set Language = "en" directly in your C# code