elevenlabs_client 0.3.0 → 0.5.0

data/lib/elevenlabs_client/endpoints/text_to_speech_with_timestamps.rb

@@ -0,0 +1,73 @@
+# frozen_string_literal: true
+
+module ElevenlabsClient
+  class TextToSpeechWithTimestamps
+    def initialize(client)
+      @client = client
+    end
+
+    # POST /v1/text-to-speech/{voice_id}/with-timestamps
+    # Generate speech from text with precise character-level timing information
+    # Documentation: https://elevenlabs.io/docs/api-reference/text-to-speech/with-timestamps
+    #
+    # @param voice_id [String] Voice ID to be used
+    # @param text [String] The text that will get converted into speech
+    # @param options [Hash] Optional TTS parameters
+    # @option options [String] :model_id Model identifier (defaults to "eleven_multilingual_v2")
+    # @option options [String] :language_code ISO 639-1 language code for text normalization
+    # @option options [Hash] :voice_settings Voice settings overriding stored settings
+    # @option options [Array<Hash>] :pronunciation_dictionary_locators Pronunciation dictionary locators (max 3)
+    # @option options [Integer] :seed Deterministic sampling seed (0-4294967295)
+    # @option options [String] :previous_text Text that came before current request
+    # @option options [String] :next_text Text that comes after current request
+    # @option options [Array<String>] :previous_request_ids Request IDs of previous samples (max 3)
+    # @option options [Array<String>] :next_request_ids Request IDs of next samples (max 3)
+    # @option options [String] :apply_text_normalization Text normalization mode ("auto", "on", "off")
+    # @option options [Boolean] :apply_language_text_normalization Language text normalization
+    # @option options [Boolean] :use_pvc_as_ivc Use IVC version instead of PVC (deprecated)
+    # @option options [Boolean] :enable_logging Enable logging (defaults to true)
+    # @option options [Integer] :optimize_streaming_latency Latency optimizations (0-4, deprecated)
+    # @option options [String] :output_format Output format (defaults to "mp3_44100_128")
+    # @return [Hash] Response containing audio_base64, alignment, and normalized_alignment
+    def generate(voice_id, text, **options)
+      # Build query parameters
+      query_params = {}
+      query_params[:enable_logging] = options[:enable_logging] unless options[:enable_logging].nil?
+      query_params[:optimize_streaming_latency] = options[:optimize_streaming_latency] if options[:optimize_streaming_latency]
+      query_params[:output_format] = options[:output_format] if options[:output_format]
+      
+      # Build endpoint with query parameters
+      endpoint = "/v1/text-to-speech/#{voice_id}/with-timestamps"
+      if query_params.any?
+        query_string = query_params.map { |k, v| "#{k}=#{v}" }.join("&")
+        endpoint += "?#{query_string}"
+      end
+      
+      # Build request body
+      request_body = { text: text }
+      
+      # Add optional body parameters
+      request_body[:model_id] = options[:model_id] if options[:model_id]
+      request_body[:language_code] = options[:language_code] if options[:language_code]
+      request_body[:voice_settings] = options[:voice_settings] if options[:voice_settings]
+      request_body[:pronunciation_dictionary_locators] = options[:pronunciation_dictionary_locators] if options[:pronunciation_dictionary_locators]
+      request_body[:seed] = options[:seed] if options[:seed]
+      request_body[:previous_text] = options[:previous_text] if options[:previous_text]
+      request_body[:next_text] = options[:next_text] if options[:next_text]
+      request_body[:previous_request_ids] = options[:previous_request_ids] if options[:previous_request_ids]
+      request_body[:next_request_ids] = options[:next_request_ids] if options[:next_request_ids]
+      request_body[:apply_text_normalization] = options[:apply_text_normalization] if options[:apply_text_normalization]
+      request_body[:apply_language_text_normalization] = options[:apply_language_text_normalization] unless options[:apply_language_text_normalization].nil?
+      request_body[:use_pvc_as_ivc] = options[:use_pvc_as_ivc] unless options[:use_pvc_as_ivc].nil?
+
+      @client.post(endpoint, request_body)
+    end
+
+    # Alias for backward compatibility
+    alias_method :text_to_speech_with_timestamps, :generate
+
+    private
+
+    attr_reader :client
+  end
+end

data/lib/elevenlabs_client/endpoints/voices.rb

@@ -133,12 +133,374 @@ module ElevenlabsClient
       false
     end
 
+    # POST /v1/similar-voices
+    # Returns a list of shared voices similar to the provided audio sample
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/similar-voices
+    #
+    # @param audio_file [IO, File] Audio file to find similar voices for
+    # @param filename [String] Original filename for the audio file
+    # @param options [Hash] Optional parameters
+    # @option options [Float] :similarity_threshold Threshold for voice similarity (0-2)
+    # @option options [Integer] :top_k Number of most similar voices to return (1-100)
+    # @return [Hash] Response containing similar voices
+    def find_similar(audio_file, filename, **options)
+      endpoint = "/v1/similar-voices"
+      
+      payload = {
+        audio_file: @client.file_part(audio_file, filename)
+      }
+      
+      payload[:similarity_threshold] = options[:similarity_threshold] if options[:similarity_threshold]
+      payload[:top_k] = options[:top_k] if options[:top_k]
+
+      @client.post_multipart(endpoint, payload)
+    end
+
+    # POST /v1/voices/add
+    # Creates a new IVC (Instant Voice Cloning) voice
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/add-voice
+    #
+    # @param name [String] Name of the voice
+    # @param audio_files [Array<IO, File>] Array of audio files for voice cloning
+    # @param filenames [Array<String>] Array of original filenames
+    # @param options [Hash] Optional parameters
+    # @option options [Boolean] :remove_background_noise Remove background noise (default: false)
+    # @option options [String] :description Description of the voice
+    # @option options [String] :labels Serialized labels dictionary
+    # @return [Hash] Response containing voice_id and requires_verification status
+    def create_ivc(name, audio_files, filenames, **options)
+      endpoint = "/v1/voices/add"
+      
+      payload = { name: name }
+      
+      # Add optional parameters
+      payload[:remove_background_noise] = options[:remove_background_noise] unless options[:remove_background_noise].nil?
+      payload[:description] = options[:description] if options[:description]
+      payload[:labels] = options[:labels] if options[:labels]
+      
+      # Add audio files
+      audio_files.each_with_index do |file, index|
+        filename = filenames[index] || "audio_#{index}.mp3"
+        payload["files[]"] = @client.file_part(file, filename)
+      end
+
+      @client.post_multipart(endpoint, payload)
+    end
+
+    # GET /v1/voices/settings/default
+    # Gets the default settings for voices
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/default-settings
+    #
+    # @return [Hash] Default voice settings
+    def get_default_settings
+      endpoint = "/v1/voices/settings/default"
+      @client.get(endpoint)
+    end
+
+    # GET /v1/voices/{voice_id}/settings
+    # Returns the settings for a specific voice
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/get-settings
+    #
+    # @param voice_id [String] Voice ID
+    # @return [Hash] Voice settings
+    def get_settings(voice_id)
+      endpoint = "/v1/voices/#{voice_id}/settings"
+      @client.get(endpoint)
+    end
+
+    # POST /v1/voices/{voice_id}/settings/edit
+    # Edit settings for a specific voice
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/edit-settings
+    #
+    # @param voice_id [String] Voice ID
+    # @param options [Hash] Voice settings to update
+    # @option options [Float] :stability Stability setting (0.0-1.0)
+    # @option options [Boolean] :use_speaker_boost Enable speaker boost
+    # @option options [Float] :similarity_boost Similarity boost setting (0.0-1.0)
+    # @option options [Float] :style Style exaggeration (0.0-1.0)
+    # @option options [Float] :speed Speed adjustment (0.25-4.0)
+    # @return [Hash] Response with status
+    def edit_settings(voice_id, **options)
+      endpoint = "/v1/voices/#{voice_id}/settings/edit"
+      
+      payload = {}
+      payload[:stability] = options[:stability] if options[:stability]
+      payload[:use_speaker_boost] = options[:use_speaker_boost] unless options[:use_speaker_boost].nil?
+      payload[:similarity_boost] = options[:similarity_boost] if options[:similarity_boost]
+      payload[:style] = options[:style] if options[:style]
+      payload[:speed] = options[:speed] if options[:speed]
+
+      @client.post(endpoint, payload)
+    end
+
+    # GET /v1/voices/{voice_id}/samples/{sample_id}/audio
+    # Returns the audio corresponding to a sample attached to a voice
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/get-sample-audio
+    #
+    # @param voice_id [String] Voice ID
+    # @param sample_id [String] Sample ID
+    # @return [String] Binary audio data
+    def get_sample_audio(voice_id, sample_id)
+      endpoint = "/v1/voices/#{voice_id}/samples/#{sample_id}/audio"
+      @client.get(endpoint)
+    end
+
+    # POST /v1/voices/pvc
+    # Creates a new PVC (Professional Voice Cloning) voice with metadata but no samples
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/create-pvc
+    #
+    # @param name [String] Name of the voice (max 100 characters)
+    # @param language [String] Language used in the samples
+    # @param options [Hash] Optional parameters
+    # @option options [String] :description Description (max 500 characters)
+    # @option options [Hash] :labels Serialized labels dictionary
+    # @return [Hash] Response containing voice_id
+    def create_pvc(name, language, **options)
+      endpoint = "/v1/voices/pvc"
+      
+      payload = {
+        name: name,
+        language: language
+      }
+      
+      payload[:description] = options[:description] if options[:description]
+      payload[:labels] = options[:labels] if options[:labels]
+
+      @client.post(endpoint, payload)
+    end
+
+    # POST /v1/voices/pvc/{voice_id}
+    # Edit PVC voice metadata
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/update-pvc
+    #
+    # @param voice_id [String] Voice ID
+    # @param options [Hash] Parameters to update
+    # @option options [String] :name New name (max 100 characters)
+    # @option options [String] :language New language
+    # @option options [String] :description New description (max 500 characters)
+    # @option options [Hash] :labels New labels dictionary
+    # @return [Hash] Response containing voice_id
+    def update_pvc(voice_id, **options)
+      endpoint = "/v1/voices/pvc/#{voice_id}"
+      
+      payload = {}
+      payload[:name] = options[:name] if options[:name]
+      payload[:language] = options[:language] if options[:language]
+      payload[:description] = options[:description] if options[:description]
+      payload[:labels] = options[:labels] if options[:labels]
+
+      @client.post(endpoint, payload)
+    end
+
+    # POST /v1/voices/pvc/{voice_id}/train
+    # Start PVC training process for a voice
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/train-pvc
+    #
+    # @param voice_id [String] Voice ID
+    # @param options [Hash] Optional parameters
+    # @option options [String] :model_id Model ID to use for conversion
+    # @return [Hash] Response with status
+    def train_pvc(voice_id, **options)
+      endpoint = "/v1/voices/pvc/#{voice_id}/train"
+      
+      payload = {}
+      payload[:model_id] = options[:model_id] if options[:model_id]
+
+      @client.post(endpoint, payload)
+    end
+
+    # POST /v1/voices/pvc/{voice_id}/samples
+    # Add audio samples to a PVC voice
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/add-pvc-samples
+    #
+    # @param voice_id [String] Voice ID
+    # @param audio_files [Array<IO, File>] Audio files for the voice
+    # @param filenames [Array<String>] Original filenames
+    # @param options [Hash] Optional parameters
+    # @option options [Boolean] :remove_background_noise Remove background noise (default: false)
+    # @return [Array<Hash>] Array of sample information
+    def add_pvc_samples(voice_id, audio_files, filenames, **options)
+      endpoint = "/v1/voices/pvc/#{voice_id}/samples"
+      
+      payload = {}
+      payload[:remove_background_noise] = options[:remove_background_noise] unless options[:remove_background_noise].nil?
+      
+      # Add audio files
+      audio_files.each_with_index do |file, index|
+        filename = filenames[index] || "audio_#{index}.mp3"
+        payload["files[]"] = @client.file_part(file, filename)
+      end
+
+      @client.post_multipart(endpoint, payload)
+    end
+
+    # POST /v1/voices/pvc/{voice_id}/samples/{sample_id}
+    # Update a PVC voice sample - apply noise removal or select speaker
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/update-pvc-sample
+    #
+    # @param voice_id [String] Voice ID
+    # @param sample_id [String] Sample ID
+    # @param options [Hash] Update parameters
+    # @option options [Boolean] :remove_background_noise Remove background noise
+    # @option options [Array<String>] :selected_speaker_ids Speaker IDs for training
+    # @option options [Integer] :trim_start_time Start time in milliseconds
+    # @option options [Integer] :trim_end_time End time in milliseconds
+    # @return [Hash] Response containing voice_id
+    def update_pvc_sample(voice_id, sample_id, **options)
+      endpoint = "/v1/voices/pvc/#{voice_id}/samples/#{sample_id}"
+      
+      payload = {}
+      payload[:remove_background_noise] = options[:remove_background_noise] unless options[:remove_background_noise].nil?
+      payload[:selected_speaker_ids] = options[:selected_speaker_ids] if options[:selected_speaker_ids]
+      payload[:trim_start_time] = options[:trim_start_time] if options[:trim_start_time]
+      payload[:trim_end_time] = options[:trim_end_time] if options[:trim_end_time]
+
+      @client.post(endpoint, payload)
+    end
+
+    # DELETE /v1/voices/pvc/{voice_id}/samples/{sample_id}
+    # Delete a sample from a PVC voice
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/delete-pvc-sample
+    #
+    # @param voice_id [String] Voice ID
+    # @param sample_id [String] Sample ID
+    # @return [Hash] Response with status
+    def delete_pvc_sample(voice_id, sample_id)
+      endpoint = "/v1/voices/pvc/#{voice_id}/samples/#{sample_id}"
+      @client.delete(endpoint)
+    end
+
+    # GET /v1/voices/pvc/{voice_id}/samples/{sample_id}/audio
+    # Retrieve voice sample audio with or without noise removal
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/get-pvc-sample-audio
+    #
+    # @param voice_id [String] Voice ID
+    # @param sample_id [String] Sample ID
+    # @param options [Hash] Optional parameters
+    # @option options [Boolean] :remove_background_noise Remove background noise (default: false)
+    # @return [Hash] Response with base64 audio data and metadata
+    def get_pvc_sample_audio(voice_id, sample_id, **options)
+      endpoint = "/v1/voices/pvc/#{voice_id}/samples/#{sample_id}/audio"
+      
+      params = {}
+      params[:remove_background_noise] = options[:remove_background_noise] unless options[:remove_background_noise].nil?
+
+      @client.get(endpoint, params)
+    end
+
+    # GET /v1/voices/pvc/{voice_id}/samples/{sample_id}/waveform
+    # Retrieve the visual waveform of a voice sample
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/get-pvc-waveform
+    #
+    # @param voice_id [String] Voice ID
+    # @param sample_id [String] Sample ID
+    # @return [Hash] Response with sample_id and visual_waveform array
+    def get_pvc_sample_waveform(voice_id, sample_id)
+      endpoint = "/v1/voices/pvc/#{voice_id}/samples/#{sample_id}/waveform"
+      @client.get(endpoint)
+    end
+
+    # GET /v1/voices/pvc/{voice_id}/samples/{sample_id}/speakers
+    # Retrieve speaker separation status and detected speakers
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/get-pvc-speakers
+    #
+    # @param voice_id [String] Voice ID
+    # @param sample_id [String] Sample ID
+    # @return [Hash] Response with separation status and speakers
+    def get_pvc_speaker_separation_status(voice_id, sample_id)
+      endpoint = "/v1/voices/pvc/#{voice_id}/samples/#{sample_id}/speakers"
+      @client.get(endpoint)
+    end
+
+    # POST /v1/voices/pvc/{voice_id}/samples/{sample_id}/separate-speakers
+    # Start speaker separation process for a sample
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/start-speaker-separation
+    #
+    # @param voice_id [String] Voice ID
+    # @param sample_id [String] Sample ID
+    # @return [Hash] Response with status
+    def start_pvc_speaker_separation(voice_id, sample_id)
+      endpoint = "/v1/voices/pvc/#{voice_id}/samples/#{sample_id}/separate-speakers"
+      @client.post(endpoint)
+    end
+
+    # GET /v1/voices/pvc/{voice_id}/samples/{sample_id}/speakers/{speaker_id}/audio
+    # Retrieve separated audio for a specific speaker
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/get-separated-speaker-audio
+    #
+    # @param voice_id [String] Voice ID
+    # @param sample_id [String] Sample ID
+    # @param speaker_id [String] Speaker ID
+    # @return [Hash] Response with base64 audio data and metadata
+    def get_pvc_separated_speaker_audio(voice_id, sample_id, speaker_id)
+      endpoint = "/v1/voices/pvc/#{voice_id}/samples/#{sample_id}/speakers/#{speaker_id}/audio"
+      @client.get(endpoint)
+    end
+
+    # POST /v1/voices/pvc/{voice_id}/verification
+    # Request manual verification for a PVC voice
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/request-pvc-verification
+    #
+    # @param voice_id [String] Voice ID
+    # @param verification_files [Array<IO, File>] Verification documents
+    # @param filenames [Array<String>] Original filenames
+    # @param options [Hash] Optional parameters
+    # @option options [String] :extra_text Extra text for verification process
+    # @return [Hash] Response with status
+    def request_pvc_verification(voice_id, verification_files, filenames, **options)
+      endpoint = "/v1/voices/pvc/#{voice_id}/verification"
+      
+      payload = {}
+      payload[:extra_text] = options[:extra_text] if options[:extra_text]
+      
+      # Add verification files
+      verification_files.each_with_index do |file, index|
+        filename = filenames[index] || "verification_#{index}.pdf"
+        payload["files[]"] = @client.file_part(file, filename)
+      end
+
+      @client.post_multipart(endpoint, payload)
+    end
+
+    # GET /v1/voices/pvc/{voice_id}/captcha
+    # Get captcha for PVC voice verification
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/get-pvc-captcha
+    #
+    # @param voice_id [String] Voice ID
+    # @return [Hash] Captcha data
+    def get_pvc_captcha(voice_id)
+      endpoint = "/v1/voices/pvc/#{voice_id}/captcha"
+      @client.get(endpoint)
+    end
+
+    # POST /v1/voices/pvc/{voice_id}/captcha
+    # Submit captcha verification for PVC voice
+    # Documentation: https://elevenlabs.io/docs/api-reference/voices/verify-pvc-captcha
+    #
+    # @param voice_id [String] Voice ID
+    # @param recording_file [IO, File] Audio recording of the user
+    # @param filename [String] Original filename for the recording
+    # @return [Hash] Response with status
+    def verify_pvc_captcha(voice_id, recording_file, filename)
+      endpoint = "/v1/voices/pvc/#{voice_id}/captcha"
+      
+      payload = {
+        recording: @client.file_part(recording_file, filename)
+      }
+
+      @client.post_multipart(endpoint, payload)
+    end
+
     # Alias methods for backward compatibility and convenience
     alias_method :get_voice, :get
     alias_method :list_voices, :list
     alias_method :create_voice, :create
     alias_method :edit_voice, :edit
     alias_method :delete_voice, :delete
+    alias_method :similar_voices, :find_similar
+    alias_method :default_settings, :get_default_settings
+    alias_method :voice_settings, :get_settings
+    alias_method :update_settings, :edit_settings
 
     private