eleven_rb 0.3.0 → 0.4.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 8f8b7a2ab5d7ebe900552e83ae3e8499545f788da927ffc60d54185464d3bbc3
-  data.tar.gz: ee1b62923b6fc88304f4c78a60f68164617fd0f46d68cca09ad1496f14dda10d
+  metadata.gz: fa71eff851a0c6b80f139e801962bceaf1bb371f6e9a0cd325c47b2ee6f4994c
+  data.tar.gz: aa640970faba75afe3cbacdc16958ee63c1f2ecfb009512cd9e8182117e16d29
 SHA512:
-  metadata.gz: 3079784a64fe6d3bff8e2c631d46763dcc6e352da16e9a7599d0ef0df7be45694c3b00279d3362f6459f7168c72f26b0e460eb80bd07239b675394979fd9cc59
-  data.tar.gz: 224530d093fdbd9b489adf6199cfd4cd19e22b2e9074bd42f59345fd0a5efcd06930c9ef9e9bfe9ab9034621c54eaf6c9443ba1b5aee91296c7365347fbcbe67
+  metadata.gz: a537ba9de014afc366c348a71613f257b6380ae784979fb42cd55522610c85661e0e34907c770642fa268051cbb91b24110e00d5e8b7305d5c913a81e04705a6
+  data.tar.gz: c1f4e236fb327b737b4e6346bf3a71bfec357d4f7f09a0e1eaf9c8889ef2251ff40b8eeb3be959e98b58e2a76ff6305e7391d64f0c33a9ce1fb38897b958f1fa

data/CHANGELOG.md

@@ -7,6 +7,20 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.4.0] - 2026-03-10
+
+### Added
+
+- Speech-to-Speech voice conversion via `client.sts.convert` (`POST /v1/speech-to-speech/{voice_id}`)
+- `Client#speech_to_speech` resource with `sts` alias
+- Accepts file paths (String) or IO objects (IO, StringIO, Tempfile) for audio input
+- Multipart upload with binary response support
+- Default model: `eleven_english_sts_v2`
+
+### Changed
+
+- `Resources::Base#post_multipart` and `HTTP::Client#post_multipart` now accept `response_type:` parameter (defaults to `:json`, backwards-compatible)
+
 ## [0.3.0] - 2026-02-08
 
 ### Added

data/README.md

@@ -4,11 +4,12 @@
 [![CI](https://github.com/webventures/eleven_rb/actions/workflows/ci.yml/badge.svg)](https://github.com/webventures/eleven_rb/actions/workflows/ci.yml)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
 
-A Ruby client for the [ElevenLabs](https://try.elevenlabs.io/qyk2j8gumrjz) Text-to-Speech, Sound Effects, and Music API.
+A Ruby client for the [ElevenLabs](https://try.elevenlabs.io/qyk2j8gumrjz) Text-to-Speech, Speech-to-Speech, Sound Effects, and Music API.
 
 ## Features
 
 - Text-to-Speech generation and streaming
+- Speech-to-Speech voice conversion
 - Sound effects generation from text descriptions
 - Music generation from prompts or composition plans
 - Voice management (list, get, create, update, delete)
@@ -88,6 +89,28 @@ File.open("output.mp3", "wb") do |file|
 end
 ```
 
+### Speech-to-Speech
+
+```ruby
+# Convert audio to a different voice
+audio = client.sts.convert("input.mp3", voice_id: "voice_id")
+audio.save_to_file("output.mp3")
+
+# With options
+audio = client.sts.convert(
+  "input.mp3",
+  voice_id: "voice_id",
+  model_id: "eleven_english_sts_v2",
+  voice_settings: { stability: 0.5, similarity_boost: 0.75 },
+  remove_background_noise: true,
+  output_format: "mp3_44100_192"
+)
+
+# From an IO object
+io = File.open("input.mp3", "rb")
+audio = client.sts.convert(io, voice_id: "voice_id")
+```
+
 ### Sound Effects
 
 ```ruby

data/lib/eleven_rb/client.rb

@@ -79,6 +79,14 @@ module ElevenRb
       @user ||= Resources::User.new(http_client)
     end
 
+    # Speech-to-speech resource
+    #
+    # @return [Resources::SpeechToSpeech]
+    def speech_to_speech
+      @speech_to_speech ||= Resources::SpeechToSpeech.new(http_client)
+    end
+    alias sts speech_to_speech
+
     # Sound effects resource
     #
     # @return [Resources::SoundEffects]

data/lib/eleven_rb/http/client.rb

@@ -49,9 +49,10 @@ module ElevenRb
       #
       # @param path [String] the API path
       # @param params [Hash] form parameters including files
-      # @return [Hash] parsed JSON response
-      def post_multipart(path, params)
-        request(:post, path, body: params, multipart: true)
+      # @param response_type [Symbol] :json or :binary
+      # @return [Hash, String] parsed JSON or binary response
+      def post_multipart(path, params, response_type: :json)
+        request(:post, path, body: params, multipart: true, response_type: response_type)
       end
 
       # Make a streaming POST request

data/lib/eleven_rb/resources/base.rb

@@ -64,9 +64,10 @@ module ElevenRb
       #
       # @param path [String]
       # @param params [Hash]
-      # @return [Hash]
-      def post_multipart(path, params)
-        http_client.post_multipart(path, params)
+      # @param response_type [Symbol] :json or :binary
+      # @return [Hash, String]
+      def post_multipart(path, params, response_type: :json)
+        http_client.post_multipart(path, params, response_type: response_type)
       end
 
       # Validate presence of a value

data/lib/eleven_rb/resources/speech_to_speech.rb

@@ -0,0 +1,94 @@
+# frozen_string_literal: true
+
+module ElevenRb
+  module Resources
+    # Speech-to-speech voice conversion resource
+    #
+    # Converts audio from one voice to another while preserving timing,
+    # cadence, and emotion. Uses ElevenLabs STS API with multipart upload.
+    #
+    # @example Convert a file
+    #   audio = client.sts.convert("input.mp3", voice_id: "abc123")
+    #   audio.save_to_file("output.mp3")
+    #
+    # @example Convert an IO object
+    #   io = File.open("input.mp3", "rb")
+    #   audio = client.sts.convert(io, voice_id: "abc123")
+    class SpeechToSpeech < Base
+      DEFAULT_MODEL = 'eleven_english_sts_v2'
+      MAX_INPUT_BYTES = 50 * 1024 * 1024 # 50 MB
+
+      # Convert speech from one voice to another
+      #
+      # @param audio_input [String, IO, Tempfile] file path or IO object of source audio
+      # @param voice_id [String] target voice ID to convert into
+      # @param model_id [String] STS model (default: eleven_english_sts_v2)
+      # @param voice_settings [Hash, nil] override voice settings (stability, similarity_boost)
+      # @param remove_background_noise [Boolean] isolate speech before conversion
+      # @param output_format [String] audio output format
+      # @param seed [Integer, nil] for reproducible results
+      # @return [Objects::Audio]
+      def convert(audio_input, voice_id:, model_id: DEFAULT_MODEL,
+                  voice_settings: nil, remove_background_noise: false,
+                  output_format: 'mp3_44100_128', seed: nil)
+        validate_presence!(voice_id, 'voice_id')
+
+        file = prepare_upload(audio_input)
+
+        params = {
+          audio: file,
+          model_id: model_id
+        }
+        params[:voice_settings] = voice_settings.to_json if voice_settings
+        params[:remove_background_noise] = remove_background_noise.to_s
+        params[:seed] = seed.to_s if seed
+
+        path = "/speech-to-speech/#{voice_id}?output_format=#{output_format}"
+        response = post_multipart(path, params, response_type: :binary)
+
+        audio = Objects::Audio.new(
+          data: response,
+          format: output_format,
+          voice_id: voice_id,
+          text: '[speech-to-speech]',
+          model_id: model_id
+        )
+
+        notify_audio_generated(audio, voice_id: voice_id, model_id: model_id)
+        audio
+      ensure
+        file&.close if file.respond_to?(:close) && audio_input.is_a?(String)
+      end
+
+      private
+
+      def notify_audio_generated(audio, voice_id:, model_id:)
+        cost_info = Objects::CostInfo.new(text: '[sts]', voice_id: voice_id, model_id: model_id)
+        http_client.config.trigger(
+          :on_audio_generated,
+          audio: audio,
+          voice_id: voice_id,
+          text: '[speech-to-speech]',
+          cost_info: cost_info.to_h
+        )
+      end
+
+      # Prepare the audio input for multipart upload
+      #
+      # @param input [String, IO, StringIO, Tempfile] file path or IO object
+      # @return [IO] file handle ready for upload
+      def prepare_upload(input)
+        case input
+        when String
+          raise Errors::ValidationError, "File not found: #{input}" unless File.exist?(input)
+
+          File.open(input, 'rb')
+        when IO, StringIO, Tempfile
+          input
+        else
+          raise ArgumentError, "Expected file path or IO object, got #{input.class}"
+        end
+      end
+    end
+  end
+end

data/lib/eleven_rb/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module ElevenRb
-  VERSION = '0.3.0'
+  VERSION = '0.4.0'
 end

data/lib/eleven_rb.rb

@@ -108,6 +108,7 @@ require_relative 'eleven_rb/resources/models'
 require_relative 'eleven_rb/resources/user'
 require_relative 'eleven_rb/resources/sound_effects'
 require_relative 'eleven_rb/resources/music'
+require_relative 'eleven_rb/resources/speech_to_speech'
 
 # High-level components
 require_relative 'eleven_rb/voice_slot_manager'

metadata

@@ -1,14 +1,13 @@
 --- !ruby/object:Gem::Specification
 name: eleven_rb
 version: !ruby/object:Gem::Version
-  version: 0.3.0
+  version: 0.4.0
 platform: ruby
 authors:
 - Web Ventures Ltd
-autorequire:
 bindir: bin
 cert_chain: []
-date: 2026-02-08 00:00:00.000000000 Z
+date: 1980-01-02 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: base64
@@ -158,6 +157,7 @@ files:
 - lib/eleven_rb/resources/models.rb
 - lib/eleven_rb/resources/music.rb
 - lib/eleven_rb/resources/sound_effects.rb
+- lib/eleven_rb/resources/speech_to_speech.rb
 - lib/eleven_rb/resources/text_to_speech.rb
 - lib/eleven_rb/resources/user.rb
 - lib/eleven_rb/resources/voice_library.rb
@@ -173,7 +173,6 @@ metadata:
   source_code_uri: https://github.com/webventures/eleven_rb
   changelog_uri: https://github.com/webventures/eleven_rb/blob/main/CHANGELOG.md
   rubygems_mfa_required: 'true'
-post_install_message:
 rdoc_options: []
 require_paths:
 - lib
@@ -188,8 +187,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.5.3
-signing_key:
+rubygems_version: 3.6.9
 specification_version: 4
 summary: Ruby client for the ElevenLabs Text-to-Speech API
 test_files: []