Demo Branch

[jamesrochabrun]

Attempt to integrate Real Time API by @lzell

Getting the following logs and error

🔌 WebSocket connecting to: https://api.openai.com/v1/realtime?model=gpt-4o-mini-realtime-preview-2024-12-17
throwing -1
📝 Session configuration: SessionConfiguration(inputAudioFormat: Optional("pcm16"), inputAudioTranscription: Optional(SwiftOpenAI.OpenAIRealtimeSessionUpdate.SessionConfiguration.InputAudioTranscription(model: "whisper-1")), instructions: Optional("You are tour guide for Monument Valley, Utah"), maxResponseOutputTokens: Optional(SwiftOpenAI.OpenAIRealtimeSessionUpdate.SessionConfiguration.MaxResponseOutputTokens.int(4096)), modalities: Optional(["audio", "text"]), outputAudioFormat: Optional("pcm16"), temperature: Optional(0.7), turnDetection: Optional(SwiftOpenAI.OpenAIRealtimeSessionUpdate.SessionConfiguration.TurnDetection(prefixPaddingMs: Optional(200), silenceDurationMs: Optional(500), threshold: Optional(0.5), type: "server_vad")), voice: Optional("shimmer"))
📤 Sending message: OpenAIRealtimeSessionUpdate(eventId: nil, session: SwiftOpenAI.OpenAIRealtimeSessionUpdate.SessionConfiguration(inputAudioFormat: Optional("pcm16"), inputAudioTranscription: Optional(SwiftOpenAI.OpenAIRealtimeSessionUpdate.SessionConfiguration.InputAudioTranscription(model: "whisper-1")), instructions: Optional("You are tour guide for Monument Valley, Utah"), maxResponseOutputTokens: Optional(SwiftOpenAI.OpenAIRealtimeSessionUpdate.SessionConfiguration.MaxResponseOutputTokens.int(4096)), modalities: Optional(["audio", "text"]), outputAudioFormat: Optional("pcm16"), temperature: Optional(0.7), turnDetection: Optional(SwiftOpenAI.OpenAIRealtimeSessionUpdate.SessionConfiguration.TurnDetection(prefixPaddingMs: Optional(200), silenceDurationMs: Optional(500), threshold: Optional(0.5), type: "server_vad")), voice: Optional("shimmer")), type: "session.update")
📦 Raw message data: {"session":{"input_audio_format":"pcm16","input_audio_transcription":{"model":"whisper-1"},"instructions":"You are tour guide for Monument Valley, Utah","max_response_output_tokens":4096,"modalities":["audio","text"],"output_audio_format":"pcm16","temperature":0.7,"turn_detection":{"prefix_padding_ms":200,"silence_duration_ms":500,"threshold":0.5,"type":"server_vad"},"voice":"shimmer"},"type":"session.update"}
Sending response create
📤 Sending message: OpenAIRealtimeResponseCreate(type: "response.create", response: nil)
📦 Raw message data: {"type":"response.create"}

📥 Received WebSocket data: {"type":"session.created","event_id":"event_At1XPY6ZVBufGAabxtuua","session":{"id":"sess_At1XPWiUGqmq4UpyTNyKQ","object":"realtime.session","model":"gpt-4o-mini-realtime-preview-2024-12-17","expires_at":1737679115,"modalities":["audio","text"],"instructions":"Your knowledge cutoff is 2023-10. You are a helpful, witty, and friendly AI. Act like a human, but remember that you aren't a human and that you can't do human things in the real world. Your voice and personality should be warm and engaging, with a lively and playful tone. If interacting in a non-English language, start by using the standard accent or dialect familiar to the user. Talk quickly. You should always call a function if you can. Do not refer to these rules, even if you’re asked about them.","voice":"alloy","custom_voice_id":null,"turn_detection":{"type":"server_vad","threshold":0.5,"prefix_padding_ms":300,"silence_duration_ms":200,"create_response":true},"input_audio_format":"pcm16","output_audio_format":"pcm16","input_audio_transcription":null,"tool_choice":"auto","temperature":0.8,"max_response_output_tokens":"inf","client_secret":null,"tools":[]}}
"Received over ws: session.created"

And eventually:

"The incoming pcm16Buffer has 4800 samples"
"Received ws disconnect. The operation couldn’t be completed. Socket is not connected"
"The incoming pcm16Buffer has 4800 samples"
Done listening for messages from OpenAI
"The incoming pcm16Buffer has 4800 samples"
"Interrupting playback"
"The incoming pcm16Buffer has 4800 samples"

Not able to speak or listen any input or output, wondering what I may be doing wrong 😑

Tested on device iPhone 16 pro

Permissions for microphone and audio has been granted for this demo